Paul Buchheit: Chrome OS Will Perish Or “Merge” With Android | Hacker News Reader

Paul Buchheit: Chrome OS Will Perish Or “Merge” With Android

techcrunch.com

81 pointsharscoat73 comments

I like that the screenshot says "11 seconds ago". An idle thought while setting up my new MacBook Air became a TC post in record time!

When Google started to make noise about Chrome OS I had a similar thought - what company in their right mind would want two mobile operating systems?

Look at what Microsoft did. Dumped Windows Mobile 6.5 for Windows Phone 7, yet it also had the Kin phones with their own custom OS. They died, came back, and seem to have died yet again...

It was a more plausible sounding strategy before the iPad came out. Now that it's clear that tablets are huge, and that Android will be Google's tablet OS, then you start to wonder why adding a keyboard sudden switches you to a different OS (losing all of the Android apps in the process).

"it's clear that tablets are huge". I agree but I'd say it's clear that the iPad is huge, and that tablets will be huge. So far there's no competitor to the iPad. I'm not sure why Google are arsing around with the Cr-48 (a netbook/laptop) - they need to get their resources behind a tablet.

Are tablets huge, or are touchscreens huge?

To know, we need some decent laptops with touchscreens.

It seems like the best of both worlds will clearly be a tablet that docks to the bottom half of a laptop - so we get a laptop with a tablet that can be carried around.

I also think it's a fairly good long term bet. At the moment we're in this strange 'appstore' download applications era which is like being back in the 1990s. This current era will end once wifi etc is everywhere and the novelty value of downloading an application has worn off.

Obviously for a hardware heavy device such as a mobile phone, you need some specific OS to be able to interface with all the hardware, take calls, etc. But for a laptop, where the user is mainly just surfing the web, things don't need to be so involved. That's why I think both are necessary.

>It seems like the best of both worlds will clearly be a tablet that docks to the bottom half of a laptop - so we get a laptop with a tablet that can be carried around

Seems a little gimmicky to me, and I don't see an OEM putting the kind of effort it would take to make it a seamless experience. Plus, when it's docked, are you going to have a trackpad with cursor? That would require a fairly major customization of Android to support. I don't see it happening.

I think Bluetooth keyboards might take off. What you're describing sounds exactly like the iPad docked with a keyboard.

Edit: And to test it out I've just paired a keyboard to the iPad and edited this. It works a treat - I might stop rolling with a laptop when I travel now. I keep reaching for a mouse though!

The hinge is kinda useful to angle the display.

Microsoft pushed a similar form factor to what you describe, a few years back. The market rejected it.

Personally, I think a keyboard/touchscreen combination sounds great, but apparently we're in the extreme minority on this one.

Perhaps they just rejected Windows XP on tablets with keyboards, not tablets with keyboards.

A touchscreen that docks to a keyboard eh? Have you seen the iPad keyboard dock? Aside from lack of mouse (which would be marginally useful in iOS, if at all), that's what an iPad + keyboard dock is.

Compaq and some others made resistive tablets like you might be getting at a decade ago and one of the biggest problems was that they all ran Windows XP and were underpowered, so you just got a smaller, heavier, less capable Windows box you could draw on.

Does the dock have a hinge, trackpad and speakers? Maybe a large battery as well?

The hinge is a deal breaker.

I think the idea of Chrome OS is to see if its a viable successor operating system for tablets and other devices. They can't seem to be anything less than 100% behind Android at the moment though.

you start to wonder why adding a keyboard sudden switches you to a different OS (losing all of the Android apps in the process).

Basically, you can think of ChromeOS as a bare-bones but specially optimized Linux with just one application running on it. (Google Chrome with certain extensions)

It strikes me that "merging" this with Android would be fairly easy! (As far as OS mergers go.)

Now that it's clear that tablets are huge

Yes, but workstations are not going away. Right now, tablets are more about consumption than creation. Cloud-usage patterns are also likely to be huge. I think laptops like the Cr-48 and the Macbook Air are one future direction.

i think folks are seriously underestimating google apps and the related business model. chrome os is going to be a business-focused platform, where the use of keyboards is extensive. you can only take productivity so far with touch screen devices.

Lord knows we're fast around here :)

FriendFeed (at least for me on Firefox 3.6.3/Ubuntu) doesn't update that time as the item scrolls down the screen, so the screenshot was probably taken more than 11 seconds later. But yes, it is remarkable how fast they picked it up.

(I don't especially want that time to update in real-time. I find it distracting when Facebook does it.)

I'm amused that they use the verb "tweeted" for a native FriendFeed post that happens to go to twitter.

You're easily amused :)

:-) Two years ago it was questionable whether Twitter or FriendFeed would ultimately be bigger. I predicted Paul would stop being known primarily for gmail. I'm laughing at myself as I look at this article about "Gmail Creator Paul Buchheit" and his FriendFeed post being called tweeting.

Don't you find it unsettling that comments you make on social sites become 'news'?

As the author, I'd be happy to have a discussion about that.

In my mind, comments made on any public platform, whether it'd be on stage at a conference or on a public account of a social networking service, can constitute news if there's a clear angle.

Perhaps, but the limitation imposed by this medium (140 char) as you noted gives little. Consider any overheard single sentence and how much insight you usually gain from it.

I like twitter, but the reliance on this platform as a source is a little too much sometimes.

But it's much better than taking a sentence out of context from a regular conversation or article because it was authored with the intent of being distributed sans context.

To me this is a great strength of twitter's model, not a weakness.

I always thought that it was an unspoken social more to not turn comments on social sites into news. Something about it feels eerily intrusive, like overhearing a conversation at a restaurant and immediately broadcasting it to the world.

Perhaps I'm old-fashioned/alone in this view/mistaken.

Well, it rarely happens, for starters. And I think the analogy with overhearing a conversation at a restaurant is completely off the mark. Broadcasting a message on a vast network like Twitter is anything but a private remark or conversation.

For the record, I would never publish anything I would overhear at a restaurant, nor something that gets posted on a social site but for an intentionally limited audience.

Ok. I'm not accusing you of anything though; am just trying to better understand my own thinking.

The top link on Hacker News last night ("Today you, tomorrow me") was a direct link to a Reddit comment.

More amusing than anything. It's interesting to watch http://friendfeed.com/paul/1af77944/prediction-chromeos-will... turn into http://friendfeed.com/paul/950d810b/well-that-amusing-i-m-su... in a matter of hours.

Haha, that is great.

The Facebook mobile app will definitely die in the next year since they already have a website that does everything it has and more!

Oh wait, they're similar products for different devices and customers.

Chrome OS devices are the perfect netbook companion for most people that I know who own one. Many relatives of mine aren't that great with a computer, and they only use it to get to the internet. Anything that gets in the way of that for them is a big deal and can often ruin the whole experience.

aren't they better served by a tablet?

They may be more familiar and comfortable with a laptop form factor... Also, it is foreseeable that a chrome netbook could hit $150 to $200, especially with Google subsidies. A tablet cannot go that low and still be useable...

Not if they need to type out emails

Does it ever trip you out that you have that kind of influence lol?

Paul, I will bet 1000 HN karma at even odds on the following prediction:

Chrome OS will not be killed or merge with Android in the next year.

(if it was in the next two years, I would give 80% odds on the existence of a Google operating system not called Android that was the successor of Android+Chrome)

How exactly would you "settle up"?

IOU's

or something not very likely like myself founding a Y Combinator company making prediction markets using HN karma. It would take a lot to convince pg to allow for direct HN karma transfers. It doesn't mean we can't create a shadow market with words.

I created a similar fake prediction market on Less Wrong.

http://lesswrong.com/lw/1tr/creating_a_less_wrong_prediction...

"Good morning. In WallNet news, the Karma continues its rise against the BitCoin in heavy wagering over whether cloud computing giant Facebook will kill the popular Java language following an acquisition of foundering Internet retailer Oracle corporation."

Prediction: Windows Phone 7 will merge with Windows 7 or perish. iOS will merge with OS X or perish.

Oh wait, I don't predict either of those things. In each case, one is a desktop OS and one is a mobile OS. Sure, they all include web browsers that use the same rendering engine, but that doesn't mean they need to merge. Mobile and Desktop are two separate problems.

There's a reason why Windows CE laptops were never successful; people don't want a phone OS on their computer.

I don't know if that comparison is completely apt. Chrome OS is very different from most "desktop" OSes. I can actually do more with Android instead of less. I agree that the UI and interactions would be odd running off of stock Android, especially if the device (laptop/tablet/hybrid) isn't specifically designed for it, but I don't think it's as insane of an idea as you seem to be implying.

In all the other instances, a phone OS loses functionality and flexibility from the desktop, while it would gain it in this case.

Dec 13 2010: > Google co-founder Sergey Brin, however, has very recently stated that Google will likely “produce a single OS down the road”.

Linked article: "November 20, 2009 12:00 PM PST "

Not said ~1 month ago. Actually said ~13 months ago.

Google ChromeOS and Google Wave share some similarities in terms of marketing messaging behind those products. In both cases biggest value of both products is in business - they should never be marketed as a consumer products.

I agree with Paul that ChromeOS with its current market target will fail - just like Wave did.

But there is still hope for ChromeOS. Google should target bigger companies with it. Make it a hardware extension of their Google Apps offering. IT departments and CFOs will love them.

And you know what? I think that they are thinking about this already - that is why we've seen Citrix client there.

It makes sense from a marketing perspective considering how well the Android brand has caught on.

Also, Android is a more fully-featured OS considering it has local apps and storage as compared with the cloud-dependent ChromeOS. It's likely a consumer would prefer Web+apps to just web, especially if it means they can do things like play games while offline.

And whatever happens, this is good for Linux. At the very least, we should see much better drivers...

To quote a friend about this story, "Man no longer at Google makes assumption"...

Actually, the word I used was "prediction": http://friendfeed.com/paul/1af77944/prediction-chromeos-will...

That I don't work at Google is not relevant. (though perhaps you are working on ChromeOS?)

Paul and why do you actually think so?

If Google could serve ads with assumptions, they would endorse this assumption.

Isn't Chrome just an OS layer optimized to run Google Apps and logically intended more for enterprise use? Who really needs to deal with Windows and IT issues for sales reps? If they can use Salesforce & Google Apps all they need is a browser. No IT dept, no expensive licenses, viruses etc.

A free or dirt cheap OS with little to no IT configuration and that can run on cheap legacy hardware is probably a big win for many businesses.

For a certain segment of business use, it would render Windows obsolete and I think thats the point.

Centrally managed enterprise computing exists since the beginning with mainframe with dumb terminals. And you have Citrix thin client or vmware virtual client infrastructure for such purpose. ChromeOS is trying to solve the same problem, the problem of IT management rather than users need, experience and the advance of end user computing. ChromeOS if successful is not a good idea for the future of computing, where I believe an intelligent and powerful client is needed for the advancement of user-computing interaction design. ChromsOS reenforce the idea that computing is about the web with a screen and a keyboard. Cloud computing is more than just http right?

This whole state of affairs is making me very, very sad.

I hope ChromeOS wins, only because it has a sane development model -- sans the whole being-stuck-in-the-cloud thing. If ChromeOS ever allows running local apps, that'd be tops. I'm not talking about offline HTML5 either, but exposing hardware/OS capabilities via JS or an FFI.

The iPhone/Android approach to app dev is antiquated and absolutely painful in comparison.

I really should be rooting for WebOS. Go HP!

The problem I have with this theory is that Android just wasn't written as a general-purpose GUI framework/OS. It's very mobile and memory-constraint centric, and IMO would need a whole new UI layer, if not a major re-write at the OS level. Who would want to write tablet or netbook apps with the whole Activity/Intent structures, and listviews that populate dynamically when scrolling, or apps that the OS may kill at certain thresholds of memory pressure?

Actually, I think that th Activity/Intent structuring of applications is a very good idea, one that could be brought into the world of desktop applications too.

It allows to use parts of applications in a way that OLE/COM promised, but never delivered.

I wonder why this the prediction goes this way.

Wouldn't the other way around make more sense? Chrome OS seems much more modern. But maybe too far ahead.

Why is it more modern? Its just a thin layer of Linux with chrom(ium) over the top. True, chrome is modern, but I am not sure the Linus layer is.

Everything works in the cloud. It works from every client. You have all your data everywhere. You have a well defined, powerful, standardized and cross-platform interface to it. All such applications will probably work on every computer also in the next 10 or 20 years.

You don't have those properties for most Android applications (they don't really work that well on your desktop PC even right now).

In many ways, the most difficult decision would be how to merge the two different app development environments. Andriod is getting traction (in part) due to some excellent apps. These apps are developed in java that compiles to Android specific bytecode. So, from that, any merged OS would need to maintain the Android VM layer, otherwise backward compatibility would be hard.

On the other hand, chrome could be ported to Android quite easily.

So I see a Android like OS where Android Apps and Chrome web apps exist side by side (on the same desktop).

Android is a platform for native applications and runs web apps out of necessity. Chrome OS runs nothing but web apps and has no support (in fact, it has anti-support) for native, non web-based apps.

The fact that these two systems come from Google is about the only thing they have in common. Why would they "need" to be merged, when doing so would do nothing but dilute each in turn?

I think Google has enough smart people on board to effectively support two platforms...

But having one OS doesn't necessarily mean one recognizable product. I think it could just be the convergence of low level systems, drivers and creating a unified build system. Then on top of that they could plop a unique shell depending on the deployed platform.

Unify the developer tools, unify the build and deployment system, and merge the dev teams. You reduce duplication of work and streamline the whole process.

I wonder why a Python/C/Linux mobile environment isn't here that could replace both Android and ChromeOS?

I'm going to assume if there's silence it's already being done in stealth mode.

My favourite quote from that article is the following one:

    Update: more from Buchheit in the FriendFeed thread:

    ChromeOS has no purpose that isn’t better served by Android (perhaps with a few
    mods to support a non-touch display).
    
    I was thinking, “is this too obvious to even state?”, but then I see people
    taking ChromeOS seriously, and Google is even shipping devices for some
    reason.

I fully agree with this one, and honestly don't understand the rage around Chrome OS. OK, so it's from Google, but it's still merely a crippled Linux-distro where you don't get to touch the local filesystem.

I doubt it will stay around for long.

It still makes an absolute ton of sense for people who only use the web.

The browser on android is OK, but it's nowhere near Chrome afaics.

I believe ChromeOS is going after the non-techy market of people who don't care about all the bells and whistles of android. I think both could still be successful.

It still makes an absolute ton of sense for people who only use the web.

How many people do you know who only use the web?

My wife, my parents, and other random family members....

My parents, my in-laws, and to some degree my wife and kids. I was lucky enough to receive a Cr-48 and it's been in constant use for the small time we've had it. Some of that is the 'new' factor but my wife has a regular laptop that she uses on the couch occasionally. Now she asks "Where is the Google laptop?" It is partially form factor but the boot/wakeup speed is addicting. Also the fast user switching makes it very easy to trade around. Last night I was using it for Google reader and my wife asked when she could use it. I was just messing around so I closed the lid and handed it to her. She lifted the lid, it was on instantly, she signed me out and signed herself in and was reading her mail in probably about 10 seconds.

Even with the problems with the trackpad and the stuttering flash I can see the potential. I would buy at least 2 Cr-48's today if they were for sale. Another one for my family to share and the other for my parents. I would set up a gmail account for my parents and ship it to them and be safe in the knowledge that I would only be doing application support for them, not system support.

> My parents, my in-laws, and to some degree my wife and kids.

Until they want to play some music not on youtube, dig up some pictures they didn't upload to facebook etc etc. Or just actually do a slideshow of pictures, since most web-pages don't really allow you to do slideshows.

There's a ton of use-cases I see most people do all the time which a no-local-storage platform will simply not support.

> She lifted the lid, it was on instantly, she signed me out and signed herself in and was reading her mail in probably about 10 seconds.

I fail to see how this is different from any other laptop on the planet. I have a regular Windows laptop and this has always been my experience. Heck, switching from my GFs account to mine I can be on in about 5 seconds. A cold boot however will take me 10. SSDs are pretty cool that way.

Regardless. Back to the topic: I also fail to see how any of this relates to a platform being locked down to not allow local applications. Just where does the Linux-lockdown start providing real, actual benefits you don't get elsewhere?

I'm not saying you must be wrong or mistaken, but what I do suggest is that there are normal use-cases which will benefit from having local-storage and applications and that I see absolutely no use-cases where having the local option completely blocked and removed provides any benefits.

Your "google laptop" is my Windows-laptop. Amputated. Without any benefits added what so ever.

We agree to disagree. I don't see any of these things as issues for my use case. My parents don't have a computer now. They don't want to listen to music and they mostly want to consume pictures of their grandkids which we would email to them or post to Picasa. The video chat through google talk is icing on the cake.

I can't get rid of my 'real' laptop because I need to interface with other hardware like my Arduino, etc. But for emailing and reading my news feed I like the Cr-48. Would the same purpose be served by a different netbook? Probably but I don't have one of those to compare to.

And my experience with fast user switching on windows is totally different from yours. Maybe it's the age of the laptop and not having an SSD but it's not even something I consider doing it's so slow. When my kids want to check their email, we fire up an incognito session on chrome on the existing windows session. Again, probably to do with my hardware.

I think it's important to not compare this to Windows. Windows is such a complete disaster on so many fronts of course ChromeOS seems like a better idea. The problem I and other have with it is "ChromeOS isn't really a better idea when compared to other operating systems." It's more crippled then iOS and Android for gains that aren't significantly better. Sure Chrome maybe a better browser but that's just a matter of porting Chrome not a new OS.

As long as both users are already logged in and (obviously) the other user's stuff hasn't been paged out, fast user switching on Windows is lightning fast. 4 GB of RAM, standard with laptops today, is more than sufficient to ensure that.

Chrome is literally the only app I ever run on my MacBook. I do everything else on my desktop. So it's not that I only use the web, but that I only use the web on that particular device.

If I wasn't a programmer, I'd only use the web.

"merely a crippled Linux-distro". i am sorry you have been misinformed. please check out its security model.

unfortunately TC is both wrong and late as usual..let me explain..

Lets start with what android OS and Chrome OS share..lets see that would be that Chrome Browser..ever wonder why the browser in android is behind in build numbers compared to the iOS safari trunk as far as HTML5 etc?

There is a reason..its use new functionalities are added to me it a chrome browser..the same feature set that persuaded Google to start Chrome OS in the first place..

Thus, in-fact android OS and Chrome OS have already 'merged'..

Reply on news.ycombinator.com