Thread: Google Chrome OS

I can see a lot of applications for this.

Have you ever been to a small-town library in Canada or the US? If they have computers, it's almost guaranteed that they'll be running their own special system. It's usually a horribly crippled Windows 2000 system that doesn't let you press the start button to launch notepad, doesn't let you open a new tab (*cough* I mean window, since IE6 doesn't have tabs). Many even disable the right mouse button. I guess it confuses people or something.

Anyway, these restrictions are usually put in place in the name of security. The library doesn't want their patrons visiting . . . undesirable sites, or downloading illegal stuff, or anything like that. The trouble is that it makes the system practically useless.

A few libraries even run all of the computers as terminals from one main server. In other words, it's like everyone has a remote connection to one computer. As you can imagine, this is unbearably slow. Close a window and watch as the system takes six seconds to paint the desktop background.

Anyway, I think that if Google does this properly, a lot of libraries would jump at an operating system which was mainly centred on the Internet. (Or they would if they had someone on staff who knew what they were doing.) It would mean minimal footprint on the individual machines. It would mean that an auto-logout system could be created which wouldn't prevent users from opening tabs. It should mean that you'd actually be able to save a web page onto a pen drive!

And if it didn't, because the system was open source, some enterprising library would hire a high school student to hack the system so that it did.

That's if Google does it properly.

That being said, it's certainly not the kind of operating system I'd want to use. Meh. We'll have to wait and see. I think that even if it's a detestable operating system, at least it will introduce a new platform for companies to release their software for other than Windows and Mac. And I'd imagine programs that run on the Linux-based Google Chrome OS wouldn't be too hard to port to other Linuxes.

Originally Posted by dwks

And I'd imagine programs that run on the Linux-based Google Chrome OS wouldn't be too hard to port to other Linuxes.

Probably there won't even be a need. If I'm guessing correctly, applications for this OS will be entirely based on Google's App Engine.

And this is where there may be a problem with this otherwise nifty concept you describe. How many companies/institutions, even if public ones, you know that are willing to have their databases hosted out of their reach? And how many will be willing to follow Goggle's App Engine TOS?

I'm more inclined to the fact they really are aiming at the consumer market. And it's their objective to promote their App Engine while competing with Microsoft Operating System and Office Suite. But while I do see a merit to the concept of a web centered computer (something I've been hearing since way back when the original BeOS team decided they wanted to go that route... and abandoned BeOS), I really cannot see any merits to this project when the idea is to strip the user of any direct relationship with their data. Heck, all it takes is my ISP to experience problems for me to not be able to use my computer at all!

Can't wait until Google start selling air, water, life...

Stick with search engines you sillies. If not, then at least just get involved in the Linux community a lot more, perhaps contributing to an existing distro or somesuch.

This is my favorite of all time: Boston Review - Richard M. Stallman: Not Free at Any Price
It's riddled with insanity. From dropping support to an humanitarian project (One Laptop per Child (OLPC): Vision)* because the operating system of a communications chip happens to be proprietary (so, give the third world children the computer, but no network), to comparing Microsoft to a company selling tobacco to children.

What scares me is not him, really. Could be a one man antics. But the fact he actually gathers a crowd (or is it a cult?) is spooky.

* Note that the computer is still completely free. And runs on Linux. And everything is bloody GPL! But no. Can't do. There's a tiny OS on a tiny chip that is proprietary. Infidels!

Originally Posted by zacs7

Can't wait until Google start selling air, water, life...

Stick with search engines you sillies. If not, then at least just get involved in the Linux community a lot more, perhaps contributing to an existing distro or somesuch.

Google has NEVER been about taking proven technology and extending it. At least as far as products they bring to consumers. Google creates NEW things. You might dislike what they do but you have to admit they have the ability to innovate.

(Google already has their own Linux "distro" you know -- those yellow Google boxes aren't exactly running Debian or RedHat...)

Originally Posted by Aparavoid

I was pretty disappointed when I read it was going to be a web based OS. But if anything it will give Linux and the open source movement more recognition which they could really use.

The only thing about Linux that doesn't fail horribly to meet reasonable expectations of quality and usability is the idea. Face it, you are talking about a basically free product that is barely holding market share against a product that costs $135.

Some of the things that cripple Linux.

1. Lack of standards. Open source is good when the community can find and fix bugs, but it is bad when it results in every tom dick and harry having their own mutually incompatible libraries, packages, and distribution.
2. Poor documentation standards. man files are obviously written by CS students, most of whom have never taken a technical writing class.
3. Lack of a functional default no brainer installation. The average user is neither qualified nor interested in making a decision about which package to install.
4. Wide support for commercial quality software packages including, heaven forbid, games.
5. Lack of a well documented standardized API. This really puts a damper on software development. Without it, a developer has to require the end user to install specific packages, which may be incompatible with some other software the user is running. While this is fine for single purpose computers, it is simply impractical for home or office computers.
6. I don't like penguins, its kinda creepy the way they waddle like that.

Originally Posted by abachler

3. Lack of a functional default no brainer installation. The average user is neither qualified nor interested in making a decision about which package to install.

I think some "newbie friendly" distributions may be coming closer to this, or even have hit the mark, but it would not be a problem anyway if the average user bought computers with OEM versions of some Linux distribution installed.

Originally Posted by abachler

4. Wide support for commercial quality software packages including, heaven forbid, games.

I think you are missing a "lack of".

Originally Posted by brewbuck

Google has NEVER been about taking proven technology and extending it. At least as far as products they bring to consumers. Google creates NEW things. You might dislike what they do but you have to admit they have the ability to innovate.

(Google already has their own Linux "distro" you know -- those yellow Google boxes aren't exactly running Debian or RedHat...)

But they're still using the Linux kernel, so all they're doing is adding the GUI.

I wouldn't mind seeing what they could do if they wrote their own OS from the kernel to the GUI. Some blend of UNIX & NT, so that they get a fast, stable OS which is very user friendly. Of course they'd have to drop this silly idea of a web only OS.

Originally Posted by cpjust

But they're still using the Linux kernel, so all they're doing is adding the GUI.

Well, I'd reckon these are modified versions of the kernel. They'd have to, in order to implement such things as their own file system.

Personally I do think they innovate. Not in the direction I'd like most of the time, let me tell you. Their search engine is old and really needs to be revisited. We need better solutions to web searching nowadays. A vertical shopping-list of search results is just not cutting it anymore. We need categorized and user parameterized results. Something that Cuil sort-of hinted at.

What I think however is very little of all this Web Applications trend. I feel it is clearly being pushed onto us, in a "if there's no demand, invent a demand" style, when in fact we don't really need it. That sort of innovation, I pass.

Originally Posted by whiteflags

>> Of course, Richard Stallman is going to be very mad when it comes out.

Why? Maybe he's off topic, but the OS is open source so I don't understand.

Actually Stallman has said explicity he "does not support Open Source":

Why Open Source misses the point of Free Software - GNU Project - Free Software Foundation (FSF)

I presume this will be just the linux kernel, not all the GNU software used by normative distros that people recognize as linux. I'll probably never use it, but I think the idea of a specialized OS is neat. It may help to spread awareness that there are more than two "operating systems" or that an OS is not (necessarily) tied to a kind of computer, for what that is worth.

If it is free and very easy to use, it is the people at MICROSOFT who will be the ones really raging.

I wonder how haiku is coming along.

ps. I think the whole discussion of "market share" linux vs. anything is silly. Part of the point of it being free, *open source* and largely volunteer created is to me is that it does not have to satisfy some group of "consumers" in order to exist. It is simply not that kind of beast. It would continue to exist with no "market share" at all; the "market share" of linux is totally irrelevant unless you are a redhat stockholder, and witness *none* of those corporations own it anyway.

I'm a long time linux user, and of the points abachler makes, I could care less about any of them and completely disagree with #2 and #5, simply because I have seen the docs made by "the competition" and they are considerably worse. Also, as an "intelligent and engaged" computer user, I think having the OS all open source completely outweighs anything else comparitively speaking. On day people will laugh at the idea that once users had so little self respect as to accept that they were not allowed to see the source code for everything on their computer. Having entire OS's that are closed source looks mind numbingly ignorant to me, and no matter how "easy to install" they are, there is simply no possible comparison because of that. Contrasting linux with FreeBSD, etc. makes sense, but certainly not with a commercial OS. Altho there are some commercialized variants, linux is fundamentally *not* a commercial product, and that is the real difference. MS does not compete with linux, it competes with Macintosh.

To my mind, the closed source, commercial OS's are by definition inferior (for my purposes) and nothing could change that. Apples and oranges. I guess Chrome will be an open source, commercial (ie, corporate controlled) OS, so now there are bananas too, which might be a step in the right direction, even if the actual form of the thing is not.

Altho I betcha they will do everything they can to subvert the restrictions in the kernel which dictate that essential components must use an open source licence. I wonder what C library they will use to build this stuff?

YES! another Linux vs rest of the world thread!

1. Lack of standards. Open source is good when the community can find and fix bugs, but it is bad when it results in every tom dick and harry having their own mutually incompatible libraries, packages, and distribution.

We call that choices. From bootloader to windowing system to browsers to office suites to web servers, you have several choices to choose from. But you are right, it does introduce incompatibilities. But if you use a more widely used distribution like Ubuntu, Fedora, Debian, Mandriva, SuSE... they usually have nicely packaged libraries and programs that just work.

2. Poor documentation standards. man files are obviously written by CS students, most of whom have never taken a technical writing class.

Apparently CS students from my university have to . At least programs fail with meaningful error messages. Not an "unknown error" and a number to call. A manual written by CS students is better than one written by marketting student IMHO.

3. Lack of a functional default no brainer installation. The average user is neither qualified nor interested in making a decision about which package to install.

I can't believe a MENSA LIFE MEMBER has trouble installing mainstream Linux distributions. For modern mainstream distributions, they are essentially boiled down to time zone selection, language selection, partitioning (there's an auto mode that resizes existing partitions, or partition the free space), and click "Install". Try a recent version of Ubuntu, SuSE, Fedora, or Mandriva before you make this statement next time (if I had time, I would record a Ubuntu installation in a VM...).

4. Wide support for commercial quality software packages including, heaven forbid, games.

That I believe is a true weakness. Not Linux's fault, though. It's a chicken and eggs problem - software vendors don't write for Linux because the userbase is small, and the userbase is small because there are no games.

5. Lack of a well documented standardized API. This really puts a damper on software development. Without it, a developer has to require the end user to install specific packages, which may be incompatible with some other software the user is running. While this is fine for single purpose computers, it is simply impractical for home or office computers.

Huh? What API? The POSIX API is pretty well documented last time I checked. And what software incompatibilities? I haven't encountered any in my ~5 years of using Linux.

6. I don't like penguins, its kinda creepy the way they waddle like that.

It's okay. I do .

And for Linux Advantages (TM) -
1. User friendliness. For example, the package management systems (APT, Yum, etc) - >10000 programs and libraries packaged in your distribution's native format. Installing one is as simple as "apt-get install firefox" (there is a cute GUI thing too if you prefer that). All dependencies are automatically resolved. And a package removal - "apt-get remove firefox" (uniform interface for package removals - have you tried uninstalling a few dozen packages in Windows? well, in Linux, you just append their names to "apt-get remove"). No reboot needed unless you are updating the Linux kernel. All your installed packages are centrally updated, too. A "apt-get upgrade" updates all your packages (compared to a few dozen reboots and mouse clicks and head scratchings and clickings to update just Windows itself from a fresh install to current). No need to check each individual sites for updates.
2. Developer friendliness. A lot of interaction between developers and users. Try to get Microsoft to implement some feature for you in Microsoft Office. I've successfully done that with a few open source developers (and have implemented a few things in my open source programs by requests of their users).
3. Customizability. You can change just about everything, easily. And even more if you are a programmer (you can implement features for yourself).
4. Choices. You have several well supported windowing systems (equivalent of the Windows Shell) to choose from, a few office suites, a few filesystems (FAT32 vs NTFS don't really count), a few browsers, a few IM clients, a few bootloaders, a few login managers...
5. Security. Malwares have never been a problem for Linux. Before you say it's because no one cares to hack Linux, no, it's because (I think) primarily of the priviledges level. Linux users don't typically run with root (admin) priviledges. Almost every Windows user does that, and it's really not their fault. Many Windows programs break when they don't have admin priv, since earlier Windows versions allow that. They write to the registry, and their installation directories. I think Windows is catching up in this regard, though. Vista's UAC and VirtualStore (a very innovative workaround) are a step in the right direction (not sure how well they work, though). For consumer computers, Windows will probably be "secure enough" that users won't need to worry about malware very soon, but I doubt mission-critical servers will switch to Windows anytime soon.

For me, I take the best of both worlds. Windows for games and entertainment, Linux for everything else. Luckily they don't try to wipe each other out when installed on the same computer. Well, Windows does, by wiping out the MBR, but that's easily fixed.

Yay cyberfish.

Originally Posted by cyberfish

For me, I take the best of both worlds. Windows for games and entertainment, Linux for everything else.

I think that pretty much covers #4. Remember, linux does not have to make money, so there really is no need at all to duplicate the windows environment so they can take over "market share". Since 99% of my computer usage falls into the "everything else" category, I stopped bothering to maintain a functional windows install a long time ago.

I would note that tho I love it, I never never recommend linux to your average joe computer user. They already have windows and Mac to choose from and I think those should do just fine there. I think once or twice someone has asked me "How hard is it to switch to linux? Will it be worth it for me?" and my answer is "It will be very hard, and no, it won't be worth it for you." I see no reason to change that situation unless you mean to see a $$ profit in it, which why bother? The things I like least about it are exactly those things which have been added in recent years to make it more "user friendly" in obtuse point and click ways that leave the underlying operating system more opaque and in that sense, *less* friendly...

Originally Posted by MK27

I wonder how haiku is coming along.

It's not. Despite what they may look it sound like.
Like so many other operating system projects out there, there's a lot of feet dragging, code being rewritten for no apparent reason other than "I can", wasting time on non-essentials and nearly complete lack of interest in partnering for the development of applications (those things that actually give a meaning to an operating system).

I mean, the bloody thing has been in pre-beta for 7 years and it doesn't even have a network stack yet. I've been keeping an eye on it for the past years. Mostly out of curiosity and because it's so cheap to just create a VM to run it. And also because, while not sharing any of its code, this is the only successor to BeOS; which I did like very much. But I don't see this going anywhere. It's just like so many other Open Source projects out there; A place to hone one's programming skills, but not real commitment whatsoever to produce a final product.

Oh.. and it's not that easy to use when comparing to what Google is proposing to do with this Chrome OS.

Originally Posted by Mario F.

I mean, the bloody thing has been in pre-beta for 7 years

!! That I didn't know. Wow. I guess there is no point in holding my breath...and they turned me down for GSOC. Ha ha. I'd have it finished by now

It's just like so many other Open Source projects out there; A place to hone one's programming skills, but not real commitment whatsoever to produce a final product.

Viewing it from a non-commercial and "participatory" stand-point, I kind of find that appealing, as long as enough of the essential things are functional. IMO there will always be people who can manage that -- so if you are like minded about what "essential" means, everything will be OK.

Eg, the thing at the top of my linux wish list is a 64 bit Skype. But since Skype is not open source, that may take forever, and still be a pain for all concerned because they have have to release like a dozen slightly different binaries for various distros. Silly!