Slashdot Mirror
Google Working on Desktop Linux
paulmac84 writes "The Register reports that Google is working on a version of Ubuntu, known internally as Goobuntu. Google has confirmed it is working on a desktop linux project, but declined to supply further details, including what the project is for. Is Google about to release this as an alternative to Windows?" Update: 02/01 00:11 GMT by SM: chrisd is the first among many to point out that this is just more fodder from the Google rumor mill and isn't something they are currently planning to release.
This has been talked about for quite a long time and even supposedly seen but what can Google do that would make this more special then any other ubuntu release/spin off?
Also a more worrying question,would you see ads incorporated?
because Google has a great track record, but I don't think Google is the right company to get all excited about when we find they're working on Linux. IBM, yes, as they have extensive knowledge and experience working with both hardware and OSes.
Google is good at... gathering and indexing information. I don't see a Google Ubuntu being much more than Ubuntu with bundled linux versions of their various apps.
Google has many fingers in many pies right now, presumably trying the
chuck-it-and-see-if-it-sticks approach. No doubt this is another of
those types of projects. If it works they'll hail it as a true MS rival,
if it doesn't it'll quietly get put down in a back room a year from now
and forgotten about.
What if Google is planning on combining this with a virtualization product so that it can be installed under Windows? Your average user is not going to be able to replace Windows. However, Google could release this as a "security zone" which people would install on top of Windows. That, combined with perhaps an improved UI and a suite of desktop software may get a more typical user to install it.
What is interesting to me is not that they are making a new version of Linux. It is the possibility of Google pack tools, being ported over to Linux. If this is true, their plithora of tools provided by google might finally start working on Linux (read: Any Linux). :)
Tools like google-Earth, desktop search(GDS), picasa are sure welcome
IF, and a big if, google released a linux distro, I wouldn't see it as an alternative to windows directly as much as an alternative to all the other already released linux distros. Indirectly it would be, at least in the initital stages. Later on it would be of course. There's a dozen or so top distros, then hundreds of smaller ones. And we also have macosx and solaris, both backed by big companies. None of them, or even collectively, have made it beyond 5% or so desktop market penetration compared to windows, even though they exist. The main problem is, none of the big hardware vendors wants to take much of a chance on any particular linux distro because they don't want to support a subset, with no guaranteees the project might be abandoned, etc. With google on the other hand, you could see the big vendors "taking a chance" on at least a parity linux offering.
Microsoft tried to beat Google at their own game with their new search systems, it would be funny if Google becomes a threat to Microsoft's age-old game of operating systems and office packages. They've got the financial backing and positive reputation.
Like Ebaums World? You'll love Shizzville
More functionality is good, right? It's obviously not going to jump out of the box and replace Windows, but it will be a good next step for damn sure. Google has a lot of power and influence now, and I, for one, am giddy that this massive corporation is so pro-linux. Google has developed some pretty radical stuff.. If you're in to linux, how can you not be excited (or at least curious) about what google + linux distro = ?
I use gentoo + fluxbox myself, but I have a lovely vmware window waiting to try it.
http://cubemonkey.net/quotes -- fortune-mod quote generator
Including the sparse failover functions? Screw Windows, I'll replace Sun and AIX!!!
Google Linux Live CD/DVD.
.) it'll install directly to your drive.
Includes:
OpenOffice.org
Firefox
Flash
Xine (with *licensed* DVD support)
Evolution
Opera
GNOME
Google Earth for Linux
Picassa for Linux
Hello for Linux
Google Desktop for Linux
Google Talk for Linux
Free!
Optionally avaliable for $25, with a combination USB flash drive/802.11g wireless card. Free access to Google Wi-Fi.
Run the live CD, it tests all your hardware, if everything is determined to be compatible (wireless, etc. .
That addresses 80% of users right there, while "saving" them from all the security hassles of Windows. Google can run an update service, and dump newer versions of these apps right on to people's systems.
Then Google can become one of the world's largest software stores, too; (like Linspire) think iTunes for Software, only have it all served by Google, and be designed to work on the Google Linux distribution.
Sure; it won't be slick as OS X. But it'll be way, way slicker than XP. And think about Google's expertise; Google is good at serving lots of customized data. No one will run a better network package management system that Google, especially if Google only has to contend with ONE "stable" version of OS. They could permit other users to access their software, but it would be unsupported; if you wanted it to guaranteed work, you'll be restricted to the Google distribution, which will be tamper resistant (think root account disabled by default, administrator only enabled for power users, requirining a specific interaction with Google (please submit a request to poweruser@gmail.com if you want your system to be unlocked).
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
So Google are producing a Ubuntu-based distribution. This isn't news as most large companies do this (hell, my last job had 10 staff and we were about to produce a Debian customisation). Google used to use a Red Hat distribution but are have been switching since at least November last year according to https://lists.dulug.duke.edu/pipermail/dulug/2005- November/016656.html.
I'll eat my dog if this ever is released to the world as a "consumer" distribution, designed to take Windows marketshare.
I guess since it wil be open source someone can discover if Google builds any nonauthorized information gathering into the system, but with all of the stuff in the news about Google would you trust an OS, for your privacy, from Google?
Well... I think the answer is simpler than it looks. I would think that revenue from Ad Sense, etc. is all secondary to this project. My thought would be that Google wants to create an OS that it will distribute for free, on the basis that unused CPU cycles are "donated" back to Google to use for its own internal processes (ie. indexing, crawling, etc.). Think about it, Google would have the worlds largest distributed parallel computer crawling and indexing away at their command all for a cost of near peanuts to what it would cost to build by purchasing all the necessary hardware.
This might be the goods. Grandma buys a Google-branded flash drive from Wal-Mart and plugs it into her crotchety old win95 box and viola! She can take it to the kids' houses and have her environment follow her around. She's got a bunch of AJAX apps running local CGI when she's offline and syncing up with her Google account when she connects to the world. The apps access OpenOffice libs to render and edit M$-docs. The jump drive appears as a data drive under M$-Ware and even has the same AJAX utilities with the same look+feel and syncing capabilities. Grandma just plugs in and goes. She can boot from the drive or just use it under the Billyware or whatever. Done right, she would neither know nor care about the OS layer. She surfs the net and sends electronic greeting cards via email. That's the extent of Grandma's computational skill set. The whole environment fits in Granny's reading glasses case. Life is good.
A full Googlized version of Ubuntu only makes sense if it was geared explicitly toward search: much like Apple's Spotlight on steroids. But that can be accomplished with an application, not a full-blown operating system. Google is not interested in building a product if it does not align with their core mission, which is search. They have no interest in destroying Microsoft completely, they do not want to get into an OS war, and they certainly don't want to start diversifying to the point where their "One Thing" becomes "One Thing In Each Market." They want to do search, and do it well.
Google also does not want to replace the infrastructure in any given market; that's too much hassle. They just want to work within it. Notice, they have no interest at all in entering the cell phone or PDA market, but they certainly make their products work very well with existing technology in those markets. I think the same will hold true of OSs: they don't want to REPLACE your OS, they just want you to search with Google FROM your OS, and hopefully click on some AdWords along the way. If that means integrating their search directly into the OS so you don't have to open a browser (a la Google Desktop) then that is a step toward their goal. Replacing the entire OS is unnecessary complexity.
My guess is that the OS is being developed exclusively for inhouse use, since Google has only confirmed its existence, not it's purpose. Everything about releasing this Goobuntu to the public is pure speculation on the Register's part. Companies roll out custom OSs for inhouse use all the time; even companies using Windows have IT departments that build their own images to propegate out to the client machines, customizing which services and programs will be available. That's a "custom OS", too.
For security, the MD5 hash of this message and sig is 09f911029d74e35bd84156c5635688c0.
If Google is working on an actual OS, and not just something to work in the background of, say, a cybercafe computer, then this may be the push Linux needs to become a major Desktop player.
Not that many people know about Linux. A lot of people know about Google, to the point that "google" has become a verb ("search for"). If someone suggests putting a Linux OS on a computer, the common man will be unsure and wary of it, knowing little about Linux, despite how much it's used regularly. However, say you want to put a new OS from Google on it, and a lot of people will open up. After all, they're used to the Google web search interface, a well made, easy-to-use thing. Surely they can make an OS, too.
If Google does it right, a lot of people will migrate. "Goobuntu" (which is a stupid name) will be a gateway drug, as it were. Those who are fine with what Google offers in its OS will stay there, while those more interested in digging deeper will move on to other distros.
Google's main hurdle is being user-friendly. Yes, yes, I'm sure you can get exactly the same result for $X_COMPONENT in Linux as you do in Windows by putting $REALLY_COMPLICATED_STRING in at the command line/terminal, but regular users will be easily confused by that. Hell, most won't even want to know about the command line/terminal. A sleek interface where most common tasks are either easy to do by the user, or done automatically, is what will push this forward.
And, if the user just has to go into the terminal line, make the commands easier to understand and more intuitive. Move instead of mv, delete instead of whatever is there now, list instead of ls, find instead of grep, help instead of man, etc. With the processing power we have these days, short (and unintuitive) commands really aren't necessary, and if anyone wants the Linux desktop to experience growth, they need to go.
I know that I, in my limited knowledge and use of Linux, routinely get frusturated having to search (I mean, man -k) again and again for a simple command.
Only time will tell, however.
Why is Google putting together a distro? Because it's the first step in a longer-term plan.
Note that Google recently hired away a Microsoft engineer who believes that Microsoft no longer knows how to ship software and believes in the web-services model. He was one of the principal architects of Hailstorm.
Here's what I see Google doing:
1. Create a usable, simple, Google distro that the masses can use for web/email/etc.
2. Market the hell out of it until they get a certain viable user base.
3. Start equipping a few thousand public libraries with a few Google Distro machines each, and monitor their usage
4. Here's the key step: in all high-bandwidth installations, CONVERT THE GOOGLE DISTRO MACHINES TO DISKLESS TERMINALS with the same UI.
5. People get used to having 'their' desktop available to them in multiple locations, spanning a disked install with networked-synched customizations to the diskless terminals.
6. The era of disk-based installs of OSs dies a well-deserved death.
7. Profit!!
If you think about it, a lot of Google's products (Gmail, the Google Toolbar) are introducing portable features. A new OS distro that they can eventually deploy as a diskless terminal version for high-bandwidth locations is the next logical step. And there will be more tears in Redmond when that happens.
Read the best of all of Slash: seenonslash.com
Its all about the Windows API. For anything to become a Windows killer, the API must be extended nearly perfectly. For machines running on x86 or AMD hardware, this is simple. The API must only run pipes from the Windows system calls to the comparable Linux system calls. When there are no system calls, the machine just runs like it should. However, other hardware adds another complexity, although that can be solved relatively efficiently (a la Rosetta). Instructions can be translated across architecture at the machine level and then execute the code natively. Obviously this would run (O)2n versus (O)n on a native machine. It wouldn't really be emulation, because the entire processor and memory structure would have to be "emulated".
Why the API, you may ask? Because the API is what gives Windows its power. Now how can I be so sure? Because most Windows users out there admit that they really dislike using their computers. But they keep coming back to them. Why? Because Windows runs the programs most users want to run. In fact, Microsoft has taken great pains to ensure the WAPI runs almost completely backward compatibly, even building in certain performance "bugs" (improving them so they run efficiently) simply so that applications that worked with Windows N will work for N++. If the popularity of an OS depended upon security, efficiency, process management, and the other technical details that we geeks care about, Windows would have died before its birth. Bill Gates' genius came from marketing, in which he persuaded all the IBM-clone companies to license Windows. Then, once a solid legion of PCs had been produced, the Windows API became ever important. Windows was always a fairly "popular" operating system, but it really took off with Windows 3.1, which led to the infamous Win95. The relative ease of use, requiring little to no DOS experience, and built in software packages, such as Works, all contributed to the overall attractiveness of the system. With the legions of developers salivating at the opportunity to pounce at all those IBM-clones, the Windows API provided the foundation of Microsoft's continued growth. What's the result? 90% (guess-timate) of the world's computers run Windows OS. Mac, Linux, and other various flavors of Unix make up the remaining 10%, along with obscure systems like OS/2 and Amiga and mainframe systems, running very old software and systems.
The WAPI isn't easy to fall. Most notably, WINE, the application for Linux and various x86 Unix boxes to run Win32 apps, is a fairly good match for Windows, but has definite bugs to be ironed out. WINE has some problems like rendering windows not completely obeying the Windows API (like QuickTime). If Google hones in on the API issue, they will be in like Flynn. To live in a post-Windows world, we need to adequately match the Windows API. We all know how far superior a Linux or Unix experience is to Windows. We also must realize that Windows is king for a reason, and to behead the king, we need to beat the king at his own game. Google is the first company to be zealous enough to really attempt a coup. Apple is too proud (though I love Apple and am running a Mac right now).
My reaction to this news about Goobuntu is, well, "WAPPPPIIIII!!!!"
Google's Dan Kegel will be speaking at SCALE 4x on the subject of Linux on the Desktop. Google will also be exhibiting at the show.
The companies you mention either did Linux as a dying gasp but didn't have enough resources (Corel*) or because it was the only way they could deal with the threat to their proprietary offerings (Novell, Sun) but their hearts really weren't in it. (Besides, none of those three had/have one-tenth the name recognition among the public that Google does.)
I think a Google Linux for the masses would be the greatest thing ever. They have the resources to make software packaging and delivery easy, they've got lots of cool apps & services they could deploy and integrate, they're smart enough to know how to make a good, clean, easy-to-use UI, they've got the resources to extensively test and then certify application compatibility (i.e., MS Office under Wine or CrossOver) and most importantly, it's a brand that everyone from a CEO to a PHB to a mail room guy knows and trusts.
What's holding back Linux adoption now? Fragmentation, and the main support options are from companies that techs swear are great but that PHBs have never heard of. Even if a manager did listen to his techs and investigate Linux, what would he see--a bunch of distros with odd names and support from a bunch of companies that come and go, none of which he's ever heard of. Google could change all that.
Ubuntu is a great distro--pretty, simple, works on lots of hardware. But it has a weird name and no particularly compelling features that would draw most Windows user. For every huge plus (no viruses!) there is an equally huge minus (my favorite old game doesn't work!). Google could change all that, too.
Basically it comes down to this: if there's one company that a) could make Linux work, b) has a compelling reason to want Linux to be a success among the masses, and c) has a name people respond positively to, Google is it. They could become a major force in both the home and the office. Google can pull it off. I really hope this rumor is true.
* Corel really could have been great. If they could have made a clean desktop and bundled NATIVE versions of Draw, PhotoPaint, and WordPerfect, it would have been awesome.
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
Yeah, that's the only thing that really makes sense. Google's strength is its network computing infrastructure. It's in the position to do what Oracle mistakenly thought they could do years ago - "the network is the computer". Bandwidth is cheap now; the only smart business decision for them in this area would be to provide a standards-based, thin client OS to connect to their online services (Google.com, Gmail, GTalk, Blogger, Maps, etc).
Think about it. What software has Google released? With the exception of software obtained through acquisitions (Picasa, Earth), it only releases web-based software (Gmail, etc) or lightweight clients to more effectively use its internet-based services (Google bar, Google talk, Google desktop).
So assuming this rumor has any merit, you'll probably see:
And hopefully:
In the short term that may mean they are targeting the internet kiosk market, but I think in the future Google expects all computers to effectively be "internet kiosks".
WHy bother with MS Office? I know it is a big standard around big business, but Open Office/Star Office can handle most things and displays most office docs. Google could help force the Open Document Standard down M$'s throat ( along with Mass. ).
I'm dreaming I know, but here's what I'd want in it:
1) PAM module to authenticate against GMail's account database.
2) The backing store for GMail is made available via WebDAV, much like Apple does with DotMac.
3a) When you log by booting a Goobuntu Live CD, your WebDAV folder is mounted as your home directory.
3b) When you log on to a system that's installed on a hard drive, it syncs it with a local disk image instead. When you log out, the synced disk image is encrypted immediately, and deleted after enough time without use.
4) I want a browser interface for some of the stuff in there, for when I can't get to a Goobuntu box. Much of this is already covered (bookmarks, mail), but I'd like more (contact list, documents).
Then, I could have a desktop machine at home, a laptop, and a Live CD. I could log in to any of them and have the same environment, with all my content ultimately stored on (and indexed by, sure) Google's servers. A buddy could come over and just use it. I could go to a buddy's house and just use his system. And so on. And if I'm at a kiosk at a conference, I can still just bring up a web interface and get some things done.
(While I'm at it, can I have SyncML too?)
With so many investors mesmerized by Google, slapping the name on a Linux distribution might finally give it enough credibility for businesses to consider a full desktop OS replacement. The choice of Ubuntu isn't an accident either...it's by far the best at supporting things like plug-and-play hardware and all the other stuff that makes Windows easy to use for people.
The best thing they can possibly do is choose a single set of applications and stick to it. No regular user wants to run or learn to use three office suites, nine media players and 50+ text editors. Google could choose one vendor, plow huge amounts of money into the project and finally force some standardization in the Linux world. That's one of the chief complaints I hear from corporate IT people about Linux...their people just want one tool to get their work done. Microsoft accomodates this by maintaining IE, Office, Media Player, etc. and making sure they play well together.
"But I do know that Microsoft offers free tech support for desktops for the first two calls."
That doesn't sound dissimilar to Ubuntu's support prices then, assuming the Microsoft definition of "free". Which isn't surprising - the cost of a call centre doesn't change much just because of your product's choice of license.
"I really have no idea if Unbuntu has on demand phone in support or if you have to wait for them to call you back."
"Response time 4 hours" apparently - we're just guessing now, but I expect there are various options from different companies.
For home users, I would have thought that calling the vendor is pretty low on their list of fun things to do anyway. Certainly if it were me, I'd want to call the local shop or repair guy. And it sounds like that's what Ubuntu is encouraging, by providing support to those shops when they need it.
"But, I'm not really a Linux guy - so I'm not all that knowledgable on the subject. I have been itching to try it out, but I really don't have the time to try and set up a machine."
The best trick I found so far was removable hard disk caddies. You can try 10 different distros in turn on your second disk until you find one that installs flawlessly and seems generally amiable, and if anything fails, or if you want to do something else for a few days, you just put the other disk back and reboot to whatever your previous OS was. Very useful when anything fundamental in Linux isn't working and you want to ask someone on the internet what to do about it. Plus the hassle you save by not fscking around with formatting partitions and dual-booting.
Alternatively, have a second PC handy. e.g. you can get a Mac Mini for about the cost of a used cheeseburger and run web browsers, MS Office, Adobe stuff, and all the rest on that while playing with Linux on your PC. It can even do the internet connection, which makes Linux much easier (no internal winmodems or broadband USB modems to make your life difficult, just tell it to use the Mac as a gateway).
"I also don't have time to configure things I really don't understand."
Me neither, but I haven't noticed any obvious differences between operating systems in this area. It's not like kcontrol isn't a widget-for-widget copy of Windows configuration for example...
I've not had to edit any textfile configs on Ubuntu, if that's what you mean. The trick is probably to dump any distro which fails to configure the graphics properly the first time -- it's not worth the frustration if an installer leaves you fighting with xfree86 just to get a graphical desktop on your first install.