One thing missing: Server Sockets
My guess is that with Server Sockets and IPv6 the whole "application in a browser" thing will finally take off.
Unlike Android, Google Chrome OS is open source. Whereas Android is coded behind closed doors — one big-name developer says it's no more open than Apple's iOS — Google's imminent browser-based operating system is built — in large part — where everyone can see it. A portion of the project remains closed — Google's boot-time- …
My guess is that with Server Sockets and IPv6 the whole "application in a browser" thing will finally take off.
I barely trust them with a search term when I'm looking for a new toaster ( use Scroogle ), let alone getting their grubby little mitts on the O/S that's supporting my day to day work. Google have proved they don't give a monkey's about our data or privacy, let's not even start with that muppet they have in charge who appears to be playing the role of Zucherberg's Uncle, i.e., treating most ordinary people with the same distaste you would a lump of dog doing's you trod in!
I will give Chrome O/S a bash, just to see what the fuss it about, but I will be watching them with very beady eyes!
is clearly an idiot.
There is a GIT repository, and a documented coding style and way of submitting patches, just like, Linux Kernel. Perhaps these same idiots consider than no Opensource too...
Is it Idiot Monday or something?
But I think they were making exagerrated statements to the way Google do all of the work in private and then dump new APIs en masse for developers to get to grips with in a month, before they encourage OEMs to start using it.
I do think it's a bit far to say it isn't "open source", but it could be more open.
Hopefully with the modularisation of the OS in "3.0", we'll see this shift.
You can download the source for the current release. That seem pretty opensource to me.
You cannot download the course to code under development. Which I (and others) don't want anyway. OK, so it stops people developing it outside fo the Google umbrella. But since Google is doing the development anyway, why is that a problem?
No-one would claim that Apple are open source. Yes, the kernel is available there too.
A Git (It's not an acronym, unless you are shouting, in which case, don't) repository and a coding style do not mean the project is "open source" at all. Where is the link for the *current* development version of Android? Can anyone contribute to that?
No'one would claim that Apple are open source because you can't download the source for OSX, can't compile your own version, can't fork it etc etc
Meanwhile with android you can
>Google do all of the work in private and then dump new APIs en masse for developers to get to grips with in a month, before they encourage OEMs to start using it.
Google don't do all the work and most OEMs are contribs. Perhaps Google think FaceBook's developers lack the skills to make a useful contribution or that they should join the OHA, like eBay for instance, if they want input at such an early stage.
Pushing a majority of their changes every time they do a new release isn't fully open source, as far as I'm concerned. There are also chunks of the code which they can't open-source, I'm told a lot of the stuff for the cellular radios falls under this. Much, MUCH more of ChromeOS is open.
"one big-name developer says it's no more open than Apple's iOS"
Which is why so many companies are porting it to their hardware and offering it... I have yet to see any 3rd party offering iOS on its hardware.
What isn't open on Android are the Google applications (maps, gmail,...), which you are free to replace with your own implementation ;)
' "one big-name developer says it's no more open than Apple's iOS"
Which is why so many companies are porting it to their hardware and offering it... I have yet to see any 3rd party offering iOS on its hardware.' So Windows 7 Phone is open source by your definition?
... but still more open than iOS.
There may be some places in the US where WiMax is all-pervasive, but here in the UK I don't think there's anywhere where you can be online all the time with a mobile device like a netbook (or a tablet, despite what Zuckerberg thinks).
Chrome OS at the moment seems far more appropriate for things like TV boxes and fixed-location web devices (refridgerators, etc.).
No disagreement that the model has issues for the average user, but there is a different way to look at it. The architectural design dictates a stateless OS with an encrypted local data cache. I could see, privacy concerns aside, that issue being a feature for certain business use cases. The impact to desktop support costs would be huge with no local software to manage, and from a security/liability standpoint the device wouldn't represent a significant security risk if lost or stolen.
Disclaimer: Yes I do understand and agree that there are significant, justifiable concerns about data privacy and Google. I am not a Google fanboy, but if you wanted to call me a ChromeOS architecture/use-case fanboy have at it.
"The question is how many people really want a machine that runs nothing but web apps."
With data caps appearing on more and more networks and the cost of data / internet access from mobile networks artificially high there needs to be a bit of a shift before i give one of these (or a dumb terminal / slate) serious consideration.
I live just outside a large UK city that is the testbed for most of BT's network design 21CN etc and i can only manage 1600 kbps down with reboots every few days as the connection just stops working. Mobile data access is even worse, intermittent 3G that drops back to GPRS / 2G.
the screenshot is from 12 months ago - probably quite a bit has changed since then - is there anything newer?
Is Android 3 source code publicly available and open for community interaction? Couldn't tell from that page but I got the impression only 2.2 was.
Android not being OSS is hardly a new discussion, it's been that way since the start...
Open Source Software is just that: software that has the source available.
Other platforms are Open Source as well, but are overseen by a steering committee (think OpenOffice), and community contributions that are not deemed worthy do not make it back into the trunk. The problem with AndroidOS is their decision of when to hit the "commit" button on their development tree. Some projects do commits after every code modification. Some, like myself, do a few days worth of work (and testing!) before a commit. Am I not open source because I choose to not release my new "feature" until I'm good and ready for public scrutiny? What if I held on to that "feature" for a few months, whilst developing other complimentary features to be pushed out at the same time? Am I now not "open" source? No. Unfortunately, however, Google takes this to the extreme and retains these new features and additions for months AND simultaneously causes incompatibilities (sometimes) with older revisions. FOSS is usually developed with a public dev tree so other developers can anticipate these incompatibilities, but a public dev tree is simply a good-will gesture from the project developers, and is NOT required to qualify as OSS. Now, "FOSS" might have a bit different take on such practices....
As far as I understand, web apps aren't completely server side, nor completely client side but a bit of the two.
I think that's not very good and we rather have to choose between the client and the server, but not a mix of them.
Web apps still require a PC and not even a light machine, they require some gears to be installed (Native Client, .NET framework and else)
The irony is that, for a large acceptance by the public, web apps should be able to work off-line since they are network dependant.
So, at this point, where's the difference with a rich client ?
"I think that's not very good and we rather have to choose between the client and the server, but not a mix of them." Why? You put heavy lifting on the backend but dump CPU intensive prettiness on the user. Perfect.
"they require some gears to be installed" What? They require a browser. Some specific uses require flash, silverlight and/or java. All these things are usually installed by default.
"where's the difference with a rich client?" Rich client is a marketing buzzword. A modern web app is viewed with a rich client, ie your browser. If you want an even better buzzword (buzzphrase?) you could say it's a "virtual rich client"! Virtual's good, right?
"We are working on the Chrome OS image - the software - but in addition to that, we are actually going and working with partners to specify components at the hardware level," Google VP of product management Sundar Pichai said last fall. "We really want software to understand the underlying hardware so we can make it much faster and more secure. It's an important part of what we're trying to do."
In 10 years I will chuckle to myself as son takes his Googlepad off to university, if I can afford it. Hmm...remind me to buy some stock...
Firstly, Android is open source. Google should let people look at development versions but they don't. Nevertheless, people can take Android, modify it how they want, branch it, build it themselves, and so on. Google Maps and a few of the apps aren't, but Google even provides an app so the user can rip those apps off their phone, put their custom Android OS on the phone, and then stick the apps back onto the custom image (which, obviously, someone would have figured out how to do anyway, but Google actually provided the tool themselves.)
Secondly... I just don't see the appeal of Chromium. I don't want a OS that's only a web browser, I run a full copy of Ubuntu on my Mini and it's just fine. And although Ubuntu 8.04 (which is what's still on there) boots a little slow, Ubuntu 10.04 boots pretty quickly.
Thirdly, I don't get this idea that something like this brings anything new security-wise, to quote "In limiting all apps to the web and confining each app to its own sandbox, Google also claims the Chrome netbook is more secure than today's machines." I just don't see it -- the apps on my Linux box all have memory protection, so they can't write to the memory owned by the OS or other apps. They are limited to where they can write to disk, and what they can read. Sounds like a sandbox to me! It could be (and probably is in theory) more secure due to the limited number of applications and programs on the system (i.e. less of a "code surface" to exploit).
Quote: Thirdly, I don't get this idea that something like this brings anything new security-wise, to quote "In limiting all apps to the web and confining each app to its own sandbox, Google also claims the Chrome netbook is more secure than today's machines." I just don't see it -- the apps on my Linux box all have memory protection, so they can't write to the memory owned by the OS or other apps. They are limited to where they can write to disk, and what they can read. Sounds like a sandbox to me! It could be (and probably is in theory) more secure due to the limited number of applications and programs on the system (i.e. less of a "code surface" to exploit). ---
Ah, but can the apps read and write each others' data? *That's* the security they're talking about. One rogue app can't go rifling thru all your other data.
I'm wary of Google's ideas about thin clients. Just because the client is "open source" doesn't mean you're going to be well off when all your data and software lives on Google's servers.
It is not realistic to run web apps yet. If I understood correctly, you will need to be online all the time.
On other hand, android is open source, but google apps are not (gmail, map, etc). However as other already pointed out, you don't have to use google apps on android.
Chrome OS isn't open source, by Google's own admission. ChromIUM OS is what they claim to be open, Chrome will be their branded version.
Don't hold your breath waiting on a real release any time soon, though - simply using their codebase to build something that works *at all* is a marathon task at the moment...