Tablets

Netbook mod performance with Joli OS

TR member dcolbert talks about his experience with Jolicloud's Joli OS mobile platform. Find out where he thinks Joli OS succeeds and where it fails.

In my last post, I talked about the killer features of the most popular mobile consumer electronic devices and wondered, "How can I get long standby time, stellar battery life, and instant-on accessibility on my Lenovo IdeaPad S10 netbook?"

My quest to assemble such a machine included taking my netbook (already modified with a 500 GB magnetic hard drive and 2.5 GB of RAM) and adding a 9 cell aftermarket battery. I also decided that, to make it fair, I should find an OS that was lighter than Win Pro 7. So, I went looking for a FOSS-based, lightweight mobile OS and finally settled on a platform called Joli OS by Jolicloud.

First off, I thought the 9 cell battery would make the device a lot bulkier, and I didn't expect it to last competitively in standby mode when compared to true mobile devices that have their OS in static ROM. But in reality, any notebook with any OS can be put into (and come out of) suspend quickly enough to be considered an "instant-on" device.

With my Lenovo, you literally open or close the LCD to resume or suspend, and that's all your iPad, Xoom, or other mobile device is doing. So, it isn't really fair to compare an iPad that spends all of its time in sleep mode to your notebook that you turn on and off.

The difference between tablets and netbooks isn't really power consumption - it's battery size. For what the iPad is driving, it has a big battery. This is partly why the iPod Touch doesn't get nearly the standby or run-time that the iPad delivers. The answer is physics, not magic. In the simplest terms, the bigger the battery, the longer it will run. The balance is how bulky and awkward it is to deliver a bigger battery.

One complaint I've heard is that extra cell batteries add weight and offset the small size of mobile devices. I suppose this is true to a certain extent, but let me just say that my work notebook is a 12" subnote Lenovo X201 with an extended battery. Even with the extended 9 cell on the S10 netbook, it's still smaller, lighter, and more portable than the X201 - so, take those arguments with a grain of salt. It doesn't add that much bulk, and it actually raises the back of the machine at a slight angle, making the device more ergonomic for typing, in my honest opinion.

As for Joli OS, I can't describe how impressed I am with the execution of this platform. It's very transparent that Linux is the foundation driving the whole thing, but it doesn't suffer from the Linux-isms that I despise. In fact, Joli OS worked exactly how a typical consumer would expect an install to work - setting up everything correctly without preachy disclaimers, additional hoops, or requiring me to acknowledge how I was disappointing the Cheeto-covered, idealist beards of FOSS geeks everywhere by using proprietary drivers. This is the philosophy I've argued the FOSS community must adopt to be truly competitive with the commercial OS platforms, and I think the ease of setting up Joli OS illustrates that I'm right.

However, the problems I experienced were, unfortunately, the fault of Joli OS. It's a really good OS, but it's not quite competitive with iOS or Android - or even WebOS or Windows Mobile 7, for that matter.

One of the disadvantages of Joli OS is that there are very few native apps, partly because it's cloud-centric. For example, the Facebook app redirects to the Facebook web page. Now, for most tasks, having a true web browser with a full keyboard and pointing device is the superior way to navigate Facebook interactively. But for Facebook consumption, a native app on a mobile device is my preferred way to quickly catch up.

Linux savvy users can also install Debian packages on Joli OS. However, I couldn't find an easy way to add icons for "side-loaded" apps to the desktop. In this sense, Joli OS still seems somewhat unfinished and very much a work-in-progress.

In addition, the advanced features of Joil OS are hard to find. For example, [Alt] [F1] brings up a terminal, and from there, you can get to a full Nautilus file manager - but the steps are difficult and poorly documented. From Nautilus, you can browse SMB/CIFS/Windows shares on a LAN, but even that was unnecessarily awkward. On the other hand, it all just worked - it easily browsed, authenticated, and accessed my shared SMB resources, which is something that Ubuntu and other mainstream full *nix distros have struggled with when I've tested similar platforms.

While the Joli OS platform wasn't perfect, it certainly could be a contender. I didn't experience all the little aggravations that were deal breakers with other Linux distros, like F-Key shortcuts for volume, Wi-Fi hardware switches that don't work without crazy hacks, and unrecognized SD slots. And if Joli OS had native app support, some ad-supported versions of killer apps, I could see it being added to the collection of OS platforms that I run.

So, I was able to create a netbook built on a lightweight, Linux-based distribution - but did it compete toe-to-toe with the big boys? When I was testing the Joli OS on my Lenovo, my iPad, and Kyros usage went down about 80%. In fact, both devices completely powered down because I forgot to charge them. Remember, instant-on is instant-on, and once you have a device that can get you online in just a few seconds, it doesn't really matter what it is.

I found that there are significant advantages to the netbook. For example, when I'm consuming Tweets on a tablet, I frequently make notes of a tweet, even re-tweet it as a sort of bookmark, so I can follow a link and explore it later on. With a netbook, I don't take this extra step - instead, I immediately follow the link.

There are also times when a touch-screen tablet format is superior, especially late at night when I'm checking Twitter and Facebook or reading an e-book in bed. With a tablet, I don't need an external light, because the keyboard is self-illuminated. It's also fairly silent - just the sound of my fingertips taping on the glass. When my wife is trying to fall asleep, she greatly appreciates both of these things.

Conceptually, my test was a success. Instant-on and long standby life are really the killer features that the iPad introduced, and those are easy to reproduce. Once you even the score in those two areas, you'll find that on many other issues, the benefits are very close to a draw. The things that are well suited to a clamshell format don't do well on a tablet, and vice versa.

About

Donovan Colbert has over 16 years of experience in the IT Industry. He's worked in help-desk, enterprise software support, systems administration and engineering, IT management, and is a regular contributor for TechRepublic. Currently, his profession...

37 comments
jeromemueller
jeromemueller

I guess I'm still missing the point. You seem to be upset that no free "linux" distribution meets your needs. I agree with you on that. I don't see why that's an occasion for name calling. And I'll apologize for the lame attempt at humor. Do you expect someone to try to convince you that you're wrong, or to defend various linux distributions for not caring if you find them useful? You can "see what I mean" all you want but you don't have a clue if I fit the mold you want. bye

dcolbert
dcolbert

So, I've really been impressed with JoliOS, and I could probably extend this into a 3rd stand-alone blog, which is maybe what I should do. For now though, just to sum up in this thread. About a week after my "testing" period, I was quite content with JoliOS and had decided to stick with it. One evening, I was in bed, finishing up, with the battery in "red" with about 1 hr 15 minutes left according to the meter - when the system shut down hard. The next morning, when I tried to reboot, it wouldn't come up. A little research indicated a problem with Grub that was best resolved by moving to Grub 2. I can find all the links again, but it is a well documented issue that happenes under a variety of situations and that multiple people have reported being victim of on Debian, Ubuntu and other Debian derived Linux distros using Grub. I do not know if it is a wider issue related to Grub, or if it is specific to Debian distros, I didn't do that much research. Multiple partitions for a mobile OS that is 64GB in size defeats the purpose of a mobile OS, Neon - just... for the record. It may be a best practice, but it is a desktop OS paradigm and not a mobile OS one. Oddly enough, since my rebuild, it hasn't been able to join my Windows network and browse my shares. I didn't change anything... this is just 100% Linux. What did I do different on the first build that I didn't this build? Not a *thing*. So why was I able to browse to shares in Nautilus on the first build but this time things are acting hokey? "It isn't LINUX, it is SAMBA that you're having a problem with"... And I'm only using SAMBA because I'm using Linux. I never have these issues with Windows machines, or OS X machines... or Android or iOS machines, for that matter. Only *nix machines. Can I fix these things? Certainly - with time and effort. Is Linux a quality, stable, reliable OS competitive with the major commercial OS vendors? An OS that an average user should ever consider? Not even close. Unfortunately, there is a *long* way to go. I'm very disappointed. JoliOS is a potentially awesome OS. It is crippled by the liabilities that are inherent in Linux and the suite of applications that run on top of Linux.

dcolbert
dcolbert

This could have happened on ANY OS... but... I've been running Win 7 on this machine for two years with no problems. As a matter of fact, the Win 7 survived fine. JUST the Joli OS part got trashed. TRASHED. Now... up side is, I uninstalled, reinstalled, and because JoliOS is JoliCloud, it resynced and copied ALL my desktop settings and apps back again. It lost my local apps that were installed as part of Gnome (outside of the JoliCloud OS portion of the kernel, for lack of a better way to describe it.) It also lost all files that I had stored locally on the JoliOS partition and hadn't backed up to cloud storage via dropbox, google docs, etc. The paradigm for recovery is a partial win, and shows the strength of the Cloud. But the weakness of the OS as a local file system is bad news. This is a documented issue with Grub, that evidently is solved by Grub2? This is news to me... but the fix looks complex and it takes off points in my mind from JoliOS's ability as a real Noob safe mobile device OS. JoliOS can't afford these kind of issues, and when they show up, they need to move to fix them quickly so that the issue is transparent from an end user perspective. At this point though, on a mobile device, you cannot trust JoliOS with irreplacable *local* data. If it is important, store it in the cloud, not on your local drive. If you lose battery, you may have to reinstall and restore from the cloud. For a lot of novice users, that might be a deal breaker. I'm going to stick with it, and hope that JoliOS improves this quickly.

Neon Samurai
Neon Samurai

"This is the philosophy I???ve argued the FOSS community must adopt to be truly competitive with the commercial OS platforms, and I think the ease of setting up Joli OS illustrates that I???m right." Cause, you knew I was going to stumbled over this one but for the join of repetition.. The faulty information you consistantly based this analysis on is an assumption that the entire FOSS community is under a single organization where all members have the primary goal of retail market dominance; this is false. You can't attribute a goal to an entire community because it's a goal some community members hold. You can't then claim "The Community is a failure and must do XYZ to achieve this goal I attribute to success." Joli OS had the short coming of a distro provided repository bursting with additional software. If Joli's goal is desktop and retail market domination then well stocked repositories is something they can provide to support the end goal. In terms of proprietary drivers, I do agree to some degree. I believe that if open source drivers don't hold up tot he proprietary ones, most users are going to need the proprietary ones. Working hardware with proprietary drivers is far better than hardware that doesn't work only because of lacking drivers. On the other side of that though, I also respect that some distributions are intentionally going for a bias towards open source software. In those cases, an opt-in step is perfectly fair; the activists get there purity by default and those of us who are more pragmatic can simply enable the unpure and get on with our own system use. Debian, ubuntu.. they both clearly state preference for free software. Debian's proprietary opt-in is fine but I wouldn't complain about Canonical focusing more on enabling new users rather than delivering a free software distro. Mandriva Free - well, it's specifically provided as Mandriva's free software focused distribution. Opting in by enabling the PLF repositories.. not a big deal.. it's an expected extra step. A warning when I choose to use proprietary drivers during setup; well duh.. Mandriva Poweruser - this one is billed as a retail distribution including proprietary parts out of the box. If it gives me long warnings about proprietary licenses or makes me enable extra repositories to get at those proprietary components then I'm going to have some solid grounds for complaint. It still comes back to realizing that the entire FOSS community does not hold, and should not be told to hold the value of providing you with a custom built distribution. You may simply not be the target user/customer for a given project or distro. Back to story time.. and cars.. cause that never gets old. The auto enthusiast community has all sorts of folks. Some are into building cars, some are into classics, some racing, some cruzing, some touring. Within the entusiast markets are active organizations and companies which support if not thrive off the enthusiasts (Car Hackers, Gearheads). Outside the community are also various auto companies; they are "outside" because they tend to be actively hostile to enthusiasts (closed computer responce codes and non-standard tools for example). So, by your measurment, the entire auto enthusiast community is failing because not all of them are actively focused on taking the F1 or Nascar top place away from the current incumbent. It doesn't matter that racing may not be an area of interest for a given entusiasts; they fail for not working towards my goal. And to suggest that the enthusiast community shouldn't universally have the goal of taking the top place away from the incumbant.. well.. that's just crazy talk.. that's just changing the argument not pointing out additional relevant information. Look, the auto enthusiast community is just going to have to change this attitude if they are ever going to compete against Porche's F1 team next year. Overall, I did like the technical details in the article and am glad to read that Joli does such a job of hardware support. It was just the bits like your "cheetos covered beard" sarcastic reference; didn't add anything to your article, only served to put your readers into a deffensive posture. And, the primary point I responded to, your dictating and attributing a single retail market metric to all of FOSSdom when it does not apply.

jeromemueller
jeromemueller

I use windows and several linux/GNU distros. I can do most of what I want with either, and for the things that are exclusive to one or the other, I use it on it's proper platform. You've got me interested in trying JoliOS, so I'll try it out to see what it does. Sometimes, I think Linux/GNU distros would be more successful if they would drop the linux from their name. They could then be judged for what they are and not what they don't intend to be. How much do you want for the EEE PC 701? I'll load Puppy Linux on it and use it as my cloud computer. Jerry

Gis Bun
Gis Bun

Another Linux distros. Now how many is that?

dford
dford

I'm inclined to agree with you Donovan. Linux would make a lot more converts if we could sort out some of the stuff you've identified. Of course, many of us would then loose interest in Linux and go off to some other OS that needed all our scripting and forum trawling abilities! - It's a catch 22 sort of thing. David

PoppaTab
PoppaTab

The only thing that holds back all Linux OS's is the fact they are linux and windows users have no clue how to run things. The true Achilles heal for linux is the fact it is Linux. Windows users will not accept much outside windows. We all know Linux will do nearly everything windows will do, but windows users are not going outside of their sphere to find out.

HiwayPilot
HiwayPilot

I've been using Jolicloud on an old HP Mini for some time now. It is actually running full blown Gnome. Most of the standard desktop apps are installed and can be run from terminal but you can also simply modify the Gnome panel across the top. Simply move the speaker ikon or the network ikon by right clicking and select 'move'. This will create an empty area on the Gnome panel where you right click, select add to panel then add the Applications menu. (and anything else you want.) I am quite pleased with this little distro and recommend everyone give it a try. Great way to recycle older hardware or see just how fast your new hardware can run.

jeromemueller
jeromemueller

I'm sure the organization the supports JoliOS will give you your money back. It's unfortunate that volunteer organizations do not have the resources to test their distributions as thoroughly as Microsoft. Few organizations do. And still bugs slip past Microsoft's testing, also. Your complaints about SAMBA sound like trashing Microsoft for a bug in an Oracle database. The organization the supports Linux does not support SAMBA. Yes, I know that SAMBA was included in your distribution, but, unlike commercial distributions, the organizations that support Linux distributions do not maintain all the products with that are included. Luckily, a *nix based distribution with full support is available. Go talk with Apple.

Neon Samurai
Neon Samurai

I've seen a file system go sideways once or twice. Most of the time, the journaling system will manage it but not always. In this case, it could be something that affects all distros using the broken version of Grub. This may be related to why Grub2 is the Debian default as of 6 Squeeze. I was a little bitter about the change due to how easy Grub's config was compared to Grub2 but I gotta say, I'm feeling a little thankful I didn't blindly flip my Debian 6 back to Grub v1. So, power cut out and dropped the machine cold and either Grub got corrupted or the journaling system couldn't recover. Booo. (I had a power cut two nights ago with my desktop; that "unexpected shutdown, scanning disks" is never fun to see.) Had to do a full install on VM then look at it through a liveCD but there's the file system; ext3 so it should be journaling. I guess the filesystem isn't going to have a chance to run recovery if it doesn't get past the boot loader though. I've had a poke around online looking for reports of corrupt grub v1 issues without any luck. Not that I don't believe it's the case, I just don't have reported details to base comments on yet. In terms of partitioning; I will stick on this point for a moment. I'm not sure how a more robust partition system would defeat the purpose of a mobile OS. The two, partition layout and overall system usage, are not really related. My 32 gig phone has dedicated storage for system OS, programs and user data. Before that, the 8 gig SD in my n810 was cut in three providing the on system OS install, my first partition install and a previous OS version with user data stored on the third partition. The Blackberry even provides multiple partition if you count the internal storage and regular use of SD added storage. I'd say partitions may not be common but it's far from a best practice that mobile devices can't adopt. Joli, as the distro provider could stamp out a partitioned image and provide the user with a more robust experience. They could even add a recovery mode that rebuilds the system partition while recognizing and leaving the user data in place. Like my phone, they could even provide a separate partition for the user added after market apps. I'd actually have more issue with eating my flash write limits with journaling rather than using a flash targeted filesystem given that Joli is targeting devices more commonly using SSD. Samba; yeah, I have my grief with it too depending on the distro. Maemo on the N810 gave me no Samba issues but something is conflicting between Debian and my NAS box after an update. I think SMB/SMB2 and PDF are very similar. PDF seems to be a documented format yet no one has ever managed to implement it like Adobe does. While SMB/SMB2 are not originally produced by Microsoft, they did adopt them and for whatever reason, no one is able to implement the protocols like Microsoft. I'd also be curious as to why something changed in your second Joli install or otherwise why it worked the first time. Can you see Samba errors in /var/log/messages, /var/log/syslog or /var/log/samba? This isn't something I'd expect an average user to dig into but given that it's not working and your not an average user, it could be worth looking at. Maybe even give you another article topic. However, this bit: "And I'm only using SAMBA because I'm using Linux. I never have these issues with Windows machines, or OS X machines... or Android or iOS machines, for that matter. Only *nix machines." Well no. Your using Samba because you have to talk to Windows machines or a NAS box setup to service Windows machines primarily. If Microsoft had there way, only Windows would be able to talk SMB/SMB2. Other OS have more native protocols but they also have to support SMB/SMB2 due to living in a Windows world. Issues also seem to be more on the samba client side versus the server daemon. In simple terms, your using Samba because your not using an NTKernel based distribution not because you are using a Linux based distribution. "OS X machines... or Android or iOS machines" For android, SMB support appears to be a third party addon so yeah.. probably no Samba there unless the app is borrowing it's libraries. osX? you might want to have another look: macname:~ username$ smb[tabkey] smbcacls smbcontrol smbd smbpasswd smbstatus smbtree smbclient smbcquotas smbget smbspool smbtar smbutil macname:~ username$ smbclient -V Version 3.0.281-apple This is an interesting discussion on Apple's future SMB/SMB2 implementation: http://www.osnews.com/story/24572/Apple_Ditches_SAMBA_in_Favour_of_Homegrown_Replacement But as of now, they are providing SMB support through Samba. Not to mention a horridly outdated version (3.5.6 is about current) so hopefully they've been maintaining it internally instead of leaving the users exposed to exploits. In the end we can agree on disappointment in Joli's current build as I can see things they could have done better also; grub2, default to jffs or another flash friendly filesystem and Samba if the issue is on the client side. We're going to remain divided on the topic of OS distributions though. I just cant' disparage an OS kernel because of issues with a distribution that happens to use it. I'm still going to recognize that distributions as separate objects due to being assembled, configured and delivered by separate organizations for separate target uses. Especially when the given issues here are not common to all distributions that happen to use the Linux kernel. "It is crippled by the liabilities that are inherent in Linux and the suite of applications that run on top of Linux." Could you clarify "inherent liabilities" in the OS kernel though? I'm thinking power management but Debian 6 (KDE) and Maemo both provide accurate battery meters and that's the only thing I can think of in the kernel which relates to this case.

Neon Samurai
Neon Samurai

A more robust install would use seporate partitions. the system partition gets reinstalled leaving your user settings and data partition untouched. I'm not sure if Joli offers such a thing or if it's purely a single partition liveCD stamper. Anyhow, I missed the description of the breakage.. what did Ansu fine? How did you manage to break the system. oh.. now I have to go skim the whole giant discussion to spot it.

AnsuGisalas
AnsuGisalas

Local data at risk, cloud data safe, a bit unexpected... but I agree that is a problem. Seems a hybrid OS is needed... something which can operate a damned disk, and still mesh up with online resources, not trying to do everything by itself. Disk Operating System, now where have I heard that before?

dcolbert
dcolbert

That is what I take away from your last argument above. Auto Enthusiasts are great people who are passionate about their technology, but no one wants to drive an auto enthusiasts car as a daily driver. And they're *weird*... like any community of people who are passionate and highly particular and picky, the auto-enthusiast crowd is full of a bunch of car-geeks who know WAY too many details about very specific features of obscure automobile facts to the point that it makes normal people feel a little uncomfortable around them. You're right, your analogy works *perfectly*. :) Listen, I think one of the places where we disagree is this... you keep talking about the end-goals. From my perspective, if Linux is going to continue to be a niche platform for hardcore hobbyists, sure, it can eek out a living there, but it is NOT of any real or meaningful consequence to the rest of the world, there. It is an oddity, a freak of nature. It might contribute ideas and directions and all kinds of interesting things here and there, but in the big picture, it is NEVER going to be DIRECTLY important. So, why even bother worrying or talking about it? You're hung up on ideals, I'm hung up on market success. I think for both ideological (hate when I *know* how to spell a word and then forget for a period of time, then remember again, just as an aside to anyone who has been reading along and has been bothered by this) and business reasons, that the Linux community as represented by say, "Linux.com" does desire these goals as an END GAME. The rhetoric and dogma that comes from the leaders in the Linux community constantly puts Linux in competition with Commercial, Closed Source operating systems. And side by side, Linux distributions generally SUCK in delivering what makes commercial closed source operating systems SO successful. As I love to point out, Mac OS X has achived more commercial market penetration in about 5 years with a FreeBSD based OS than all Linux distros combined have achived in 15. You don't have to be a Microsoft Fanboy to see how Linux *fails* in this regard. It isn't baiting, it isn't trolling. It is pointing out truths. Your philosophy may be more open, genuine, broad, and pragmatic about where Linux can define "success" outside of the traditional concept of what that would mean. I'm totally cool with that. But by the criteria for competition and success that major movers and drivers in the Linux community have defined, Linux in general is doing a very poor job of competing. It is auto-enthusiast kit cars being pushed as alternatives to conservative grocery getter sedans. No surprise, these kit cars fail in the role as soccer mom mini-vans... time and time again. Ultimately, if you want a viable alternative to the commercial desktop OS platforms, then you need to embrace OPT-OUT for closed source features that are EXPECTED. Look at it this way. If you include the BEST drivers and other components by DEFAULT, even if they are closed source and proprietary, you're going to satisfy the least competent users. At the same time, the users most likely to have ideological problems with closed source and proprietary, non-free components in their OS, are generally going to be the users most capable of disabling those features and opting OUT of their own accord. Does it NOT make more sense to use the best, in all cases, and allow those who understand best what they're doing to opt out over ideological reasons, rather than using the worst, and forcing those who know the least to become frustrated with a substandard experience that they don't understand? On the other side of that though, I also respect that some distributions are intentionally going for a bias towards open source software. In those cases, an opt-in step is perfectly fair; the activists get there purity by default and those of us who are more pragmatic can simply enable the unpure and get on with our own system use. I disagree, the activists put their ideals above the experience of the least able users, and hurt the reputation of Linux in the eyes of those least able users in the process. It perpetuates a reputation and atmosphere of elitism, exclusion and snobbery among the Linux community. It comes across as preachy and self-righteous. It is refreshing to see Joli-OS dismissing this counter-productive approach.

Neon Samurai
Neon Samurai

I agree. Much of the confusion is the focus on a specific kernel that happens to be in use versus the actual software platform that is layered on top of it. The distribution is the consumer product level and that's where branding needs to be differentiating. People don't care that Tomtom has a Linux kernel, they care that they have a GPS that works and, if awayre, that it happens to be a Tomtom unit. People don't care what kernel is behind Android. They primarily car that the device is branded as an Android or they only know that "it's a phone but not the apple one" and that they can add apps on. Debian doesn't magically stop being Debian when you don't use the Linux kernel. It's a software platform that happens to use the Linux kernel not a Linux kernel that happens to have a software platform stacked on top.

dcolbert
dcolbert

But Debian based distros are clearly becoming the dominant platform among Linux... and that is a GOOD thing. That indicates market forces at work, regardless of if Linux idealists think market forces should affect Linux or not. And the Debian based distros keep getting better and better - and this one is GOOD. Did I mention I gave Easy Peasy a try? It was OK... I was afraid I'd really like it, too - or it would be remarkably similar to JoliCloud. It wasn't, and I didn't. I actually *like* JoliOS. It *is* a Linux distro worth checking out... IF: You're interested in a light, simple mobile OS type platform presented in a visual way very similar to Android and iOS. And - that kind of approach works well on a Netbook. Having a full blown desktop OS on a Netbook always seemed a little bit like a square peg in a round hole. The JoliOS feels *right* on my Netbook. What would be AWESOME would be a Netbook with touch screen and the Joli OS integrated in a ROM image similar to your typical Android handset. But you know what, Gis - when I wrote this article - the thought you state here crossed my mind. In particular I thought of Neon Samuria and how he has suggested so many different distributions over the last few years to my complaints, and how time and time again, he never really seemed to get how the very act of saying, "Well, you're using the WRONG Linux, you should try THIS one" only made me dislike Linux more. Because here I am, finally having found a Linux I like, saying almost the exact same thing. But keep in mind, I'm saying it from the position of anything BUT a Linux advocate. If you've tried all the other Linux variants, Puppy, Mint, Ubuntu, Debian... and they've all left you unimpressed, and you're INTERESTED in a mobile device type OS that is going to work well with most hardware without a lot of effort right out of the gate... THEN... Trying JoliOS couldn't *hurt*. I'm not saying it will be the end-all-be-all OS platform that ends your OS searching for all eternity... But it doesn't suck. Coming from me - about something built on the penguin kernel... that is a high compliment. :)

dcolbert
dcolbert

One of my favorite examples is that I was one of those dark, angsty, troubled teens in ripped denim jeans with a marlboro hanging from my lips hanging outside of the arcade. I was only into the music or scene that was just breaking, still underground, underproduced and raw. The minute it became mainstream, it was uncool, and I moved on - along with all the other kids like me who were "early adopters" of fashion and music. There is absolutely an *obvious* bit of this quality to the Linux community - and the minute Linux goes mainstream and becomes accepted by massive amounts of people, Linux will have as much credibility as Lars Ulrich having lunch with lawyers from the RIAA at a Michelin rated restaurant. "Dude, Linux used to be way rad, but Linus totally sold out man... now I'm into NewOS... it is so rough you have to write your own code just to get a directory listing. Totally raw and hardcore..." *cue sounds of bong rips... er, I mean Cheetos wrappers breaking open*...

dcolbert
dcolbert

That seems to get it right. JoliOS is clearly interested in appealing to a CONSUMER audience of mobile desktop device users - and they're doing the things that need to be done to be competitive. At least, some of the things. They've got a long, tough road ahead of them - though, and it seems like MAJOR players are aligning behind other mobile Linux distros. Which is puzzling. I've tried Moblin, MeeGo and Easy Peasy. Easy Peasy is the best of these 3, and it still doesn't have nearly the potential that JoliOS has. Intel, Nokia, and the other major vendors exploring these alternatives are wasting their time and headed down dead-ends. If all of these vendors got behind JoliOS and stopped their own fruitless solutions, they could come up with something that could be a significant challenge to the incumbents (Android and iOS, as well as Win and Mac Classic). My biggest concern is that JoliOS will fail not because it is the worst solution. It isn't. It is the best I've seen Linux produce other than Android. I'm afraid that JoliOS will fail simply because they won't be able to compete with the big guns and big marketing coming to this segment with their own, inferior solutions. The JoliCloud team needs to begin forging lots of close, active relationships with lots of A-teir vendors, manufacturers and other key players quickly - if they want to survive. I think the shame of it is that they probably won't. I think the technology they develop here may end up in something else that does make it eventually, though.

jslozier
jslozier

What hurts Linux is that there are very very few computers that come with it preinstalled. Even most white box dealers install Windows unless asked not to. Most users are very scared of installing or reinstalling their OS, it sounds very dangerous, which it can be if you do not backup your data. If you had a few percent or more come with Linux preinstalled it might see a sharp rise in popularity.

dcolbert
dcolbert

This is a Linux community cop-out. If you can offer a free product that meets or exceeds what a paid product delivers, consumers will pick the free product every time. Just ask Netscape. If you've got a free product and you can't PAY people to take it, there is something wrong with that product. The Linux community has been playing the "Yeah, it is lamer Windoze Luzers" card for a decade now. That is hogwash. If Windows users won't adopt Linux - then there is something wrong with LINUX. Absolutely. We can dance around that elephant in the room all day - but that is the truth of it. IE killed Netscape because IE was an equal or superior product for FREE. Linux has failed to make a statistical dent in the number of Windows and Mac OS instances on the desktop. Because Linux is inferior at any price, including free. Linux advocates do not like it when they are confronted with this truth. Here is another way to put it. You claim that Windows users will not accept much outside Windows. Well, OS X has had huge - and although a good deal of that has been converted Linux users, and loyal Classic OS users, a *large* number of those users have been Windows users who converted. So, you're wrong. Windows users will accept a *nix based OS that is radically different than Windows. They just haven't accepted any based on *Linux*. So, it *isn't* the users, the only thing that remains is the OS. I've maintained for awhile now that the list of what those things are is realatively simple and easy to address. Package Management Headaches: Debian's package management went a LOOONG way to address this. Before that, trying to install an app on Linux was simply Yabba-Dabba-Doo stone-aged technology. They might as well just have required you to load your binaries by punch-card reader. Still, it should ALWAYS be as easy as Android's .apk or Debian's .deb format. Clicking Install.exe and clicking YES several times and watching the status bar *is* how software should install. Maybe a couple of chicken-switches where the system says, "Are you SURE and do you have permission". Out of the box, no hassle installs: This is another one where Linux users always make this claim that Windows is just as bad as Linux and that Windows has the advantage of being a default OEM system that is pre-installed. I disagree. Sure, you might occasionally have a driver disk you have to put in, but in general, it is all Plug and Play in the Windows world. That still isn't the case for Linux in general - and part of that is philosophy - Linux idealists who oversee distributions won't automatically enable closed-source drivers, even when they exist, instead opting to install inferior reverse engineered FOSS drivers. Which is fine. Make a stand. But that stand is going to alienate the majority of the potential market for users. That isn't the fault of the user, it is the fault of the distributions. Now, you may have a hassle getting your v.32 bis ISA modem to be recognized under Win 7 - but in general, it is going to find all your hardware. It is also going to recognize your video card and your monitor and figure out the correct sync, refresh and resolution and set those accordingly. It is also going to install your WiFi and be able to connect easily to your WiFi router using WPA2 encryption. Flash will install correctly. Fonts will install correctly. And if you REALLY mess things up, generally it is going to be something that some Windows expert friend of yours can fix over a couple of beers and without starting from scratch. Linux simply doesn't compete in these regards. Lots of readers call me a Microsoft Shill and claim I have an irrational dislike for Linux. This isn't true at all. I point out *truths* about the Linux community that are not popular - that the Linux community would rather ignore. I gleefully tip every sacred cow in the temple of the penguin, at every opportunity. So I am not *shocked* that many members of the Linux community think I dislike Linux - but they're still wrong. I just think they're full of it, and I'm calling them on their hogwash where I find it. (Which is almost anywhere you see a Linux advocate writing a response...) JoliOS goes a *heckuva* long way to address the issues, complaints and shortcomings of Linux that I discuss above. There are some issues, and if I didn't know quite a bit about *nix, I'd be frustrated by some of the sacrifices that I would be making. A great example of where a thing should be easier on JoliOS... I was so impressed with JoliOS on my S10, I decided to try it on my Eee PC 701. There were a couple big problems. The SSD flips up a "drive use outside of parameters" Smart error that is well documented on Debian/Ubuntu distros. Trying to fix that required changing flags in the etc/fstab from vi, gedit, etc. When I followed the advice, it rendered my system unmountable. I had to boot from an Ubuntu live CD, mount the file system, get to the fstab, and remove the entires. The (s)kip and (m)anual recover keys on JoliOS did bunk-all. Average user making the switch would have been done and back to Windows XP at the Smart errors. Those who got further and bungled their system probably wouldn't have been able to recover it. Linux has to tighten up their game in this regard. They're behind the curve that Microsoft and Apple are leading in this area. The second issue was that YouTube videos were running slow - choppy and not synced to audio. The EEE PC 701 is factory default at 640Mhz with a 70 Mhz system bus, but the native speed is 900/100. Running a Win32, there is a simple utility that rides in the task bar that allows you to overclock on demand with one click. For Linux, this isn't the case. It is a fractured *mess*. Depending on which kernel you have and which distro, you may have to run scripts, or you may be able to run one-click executable solutions, or you may have no luck what-so-ever. Now, the take away for me is that I had a party interested in the Eee PC 701, but the choppy video cooled their interest. I know that with XP on this machine, without overclocking, YouTube works fine. But I don't have an XP license to include. So I looked into overclocking it, which would have satisfied the buyer. No dice. JoliOS is built on an older Ubuntu release and kernel - so it doesn't support the on-the-fly single-click overclock utilities. It isn't a complete Ubuntu platform, so the scripted solutions don't run correctly and break, too. I *could* fix this, I know what steps are required, but it is way too much work for the payoff. These are relatively minor quibbles, and the fact is that JoliOS shows the progress of the Linux community to addressing the majority of significant issues that have plagued Linux distributions for the last several years. They're on the right track. Better, more efficient codec packs, closed source ones, might be the solution to not being competitive on rendering streaming video sources. The good news is, it looks like JoliOS isn't hung up on idealism in situations like these. That is an attitude that is long overdue in the Linux community - *if* it wants to compete in this space. I hope to see great things from JoliOS.

jeb.hoge
jeb.hoge

That trick in Gnome is just what I needed and I wasn't quite clever enough to figure it out myself. Thanks!

dcolbert
dcolbert

JoliOS, in many ways, delivers on a LOT of the promises that I feel Linux has made but failed to deliver in the past. It does revive old equipment - but it also supports relatively new equipment very well, too. It doesn't compromise a lot for what it is trying to do. I've learned a lot more about the platform, too. If I had it to do over again, I would have researched the platform for 4-8 weeks before I wrote about it. It is fairly poorly documented, so a lot of things you have to find or figure out for yourself - and it isn't as always as intuitive as an iOS or Android platform device to stumble your way through learning how to operate it. It is built on an older Ubuntu distro and kernel... and that leads to some frustrations. I've got a lot more to say about this platform. It has the most promise of any Linux distro I've ever played with - and it has the interest of my industry peers who've seen me using it. "What OS is that?" I'm in a hurry tonight, but I'll be checking back in here and elaborating on my experiences since I submitted this blog. Ironic after so many people suggesting so many distros to me over the years, the one that I've liked the best, I stumbled across by accident. JoliOS has the foundation that MeeGo, Moblin, the Ubuntu Netbook Remix, all lack. All they have to do is get some big guns behind them supporting them, get pre-bundled onto more hardware - and they might have a decent shot in this space.

dcolbert
dcolbert

I'm sure the organization the supports JoliOS will give you your money back. Way to epic fail. You do not know enough about my history and you've set yourself up to be an effigy of everything I've ever claimed is wrong with the Linux community. Your quick, ill thought out statement here robs the credibility of countless Linux advocates who have engaged in meaningful, thoughtful, honest and heartfelt discussion with me on these topics. They're not going to speak up and admit this, but I guarantee you more than one loyal reader and Linux advocate here on TR read this post of yours and did a facepalm, well aware of what my response to this was going to be. The organization that supports JoliOS, if they can't satisfy *me*, will likely wither and die on the vine, making no useful contribution to ANYTHING. The fact that they produced their product as volunteers and gave it away to me as free is absolutely MOOT - it has no relevance. you're making excuses. Ultimately, no one who might be interested in putting JoliOS or any other Linus distribution on their PC is interested in excuses. Outside of Linux idealists who put up with Linux quirks because of philosophical reasons, the rest of the public is only concerned in what works, and how well it works - not if it is free or paid, open or closed, a kernel and platform or an OS, or any other split hair. And of Linux idealists, there aren't enough to go around. So JoliOS has the opportunity to idealize itself and volunteer effort itself right out of existence. The organization that supports JoliOS needs to make sure that it captures me and KEEPS me and doesn't let me down and send me back to Microsoft. If they don't have the resources... wait... wait... the Many Eyes Model is supposed to make their bug checking SUPERIOR to competitors like Microsoft. Man... this is the problem with the Linux community... you're all coming from different directions, and you don't even agree with one another... But assuming that I was RIGHT in *that* argument (which you also probably missed) - and Microsoft has more resources to make a more reliable OS because they are a closed, for-profit commercial entity... That still doesn't matter. I'd rather PAY $150 for a reliable commercial OS that doesn't eat my data than get a free one that can't go down gracelessly from a power outage without eating all the data on my drive. Almost everyone else in the world would AGREE with me on this. (Almost everyone. Those who would disagree can easily be discovered by looking for overweight guys with greasy hair pulled back in pony-tails and cheetos crumbs in their beards)... Really, this isn't so much addressed at you, poster - as toward the rest of the Linux crowd regulars I generally argue with on these threads... it is a "see what I mean" moment for me, as I kind of wave my hands, palms up, in your general direction... They know what I mean... even if this response has left you somewhat in the dark.

dcolbert
dcolbert

I'll start you off with a couple of links related to the issue that should expand on it for you: http://www.linuxquestions.org/questions/linux-newbie-8/grub-installed-boot-from-hdd-but-no-loaded-kernel-827551/ and http://ubuntuforums.org/showthread.php?t=1432486 Not exact matches for my issue, but same errors, same symptoms, same kind of results. the grub error: no loaded kernel error is the common thread. I'm going to give you the point on partitioning. It could have saved my local files if they had /home on a different part/slice than the OS. I still would have lost local installed native apps - but at least my local user data would have been preserved. If there is logic for NOT doing that way, I'm not sure what it is, but I would say this much - I wouldn't rule out that there IS a reason for JoliOS being designed this way to accomodate mobile oriented devices and easy user experience, perhaps? There may be something that isn't transparently clearly on the surface. I understand Mac OS X uses Samba... but Apple is doing something that makes it work flawlessly, without the little random annoyances I see on Linux (and Unix). What that is, I don't know... but let's capture a window of time, say since 2007 - I've built and used multiple boxes that use Samba, including Mac machines and Android (Android is using Samba, I am certain... although it may be integrated into particular Android utilities that are downloaded from the market)... but only pure Linux machines have given me fits, and on multiple occassions. It is more than a statistical coincidence. I mean, I agree with a lot of the finger details you address in this part of the post (issues are with client side, not SAMBA server... true... unfortunately)... but my observations are true, too. I'll compromise with you on the end. I'll recognize the distributions as separate, but all sucking equally in many of the same ways. ;)

dcolbert
dcolbert

I meant to put a link in the other thread, and life, the universe and everything conspired against me to keep me from doing so. Technology was not my friend at all, yesterday, so I got drunk and ignored technology, and then things went much better. It is a huge disappointment to me. I've reloaded and I still plan on using JoliOS, but it really dampens my enthusiasm for the product. The fact that it is a well documented Grub/Ubuntu issue bothers me, too. We all know where my opinions on these kind of issues are - so I won't revisit them so we can argue them in detail... but I'll just leave it at this - JoliOS could be a very competitive product, but in this state, I'd say it isn't ready for prime time.

Neon Samurai
Neon Samurai

" Ultimately, if you want a viable alternative to the commercial desktop OS platforms, then you need to embrace OPT-OUT for closed source features that are EXPECTED. " Well, yes. You'll notice that I've also frequently pointed out that if Canonical really wants to deliver a new user friendly distribution, they should reconsider the preferance for free software in the default build. Debian, sure, they outright state that free software is preferable and the opt-in is fair given how easy it is. Canonical though? They are indeed going after the new user desktop market and things like hardware support should be valued over software source availability. I'd go so far as to ask why the heck Canonical isn't doing a proper rolling distro; an average user can stuff in a disk and press "upgrade" much easier than a scrub and install. (I'm actually not sure how canonical does for upgrades between release versions but I'm guessing it's using the same dist-upgrade inherited from Debian.) But this is all part of a healthy competitive market. The activist distributions put more value on free software than on enabling the end user by default; that's fine.. let new users adopt the distributions from the more pragmatic who don't sacrifice the user experience over software licenses. The point returns again. Your deciding that the activists need to go after the new user market. If the activist's core value is new user conversions then sure. If the activist's core value is source availability then not so much; let them strive for there purity of distribution. Reward those who instead strive for you as an end user. "they are not doing it right because they are not striving for my goals" - yeah.. so.. why not instead focus on that distro over there which is indeed focusing on your needs? That's not a "your doing it wrong, pick my favorit distro instead".. that's applying the same buying process that we use in every other product category. Car doesn't fit our needs or price range; don't buy it. Appliance doesn't fit in your kitchen; buy one that does. Don't like mint flavored toothpaste; consider a cinamin flavor or different brand all together. When is it accepted natural decision making in other product categories but dismissive geek snobbery suddenly because the topic is software platforms instead of raisin toast brands? I like that Mint adds the polish and hardware support that Canonical leaves out. Joli is doing a better job of end user experience; fantastic. I just can't demand that the entire community and all distros within it adhere to what Mint or Joli is doing. I can't dictate market economic goals to the entire community. " if Linux is going to continue to be a niche platform for hardcore hobbyists, sure, it can eek out a living there " "Linux" is just trying to be an OS kernel and it's doing a pretty solid job of that goal. If you want to stick with market metrics; it's been a staggering success. In reality, I'm not seeing "replace all other OS kernels in use" as a primary goal over at kernel.org. It seems they are more interested in developing an OS kernel rather than measuring dicks through market percentages. If market domination is not a project goal, how do you draw any relevant outcome from measuring it's success/failure against market percentages instead of it's actual goals? Now, at the distribution level, we have something that can actually be discussed. Some distributions are indeed happy being in a niche. Other distributions are indeed striving for greater market dominance. One must recognize the difference. You can't decry Debian for not taking market share from Microsoft hand over fist; it's not there primary goal. You can decry Canonical because dominating the desktop space across nerds and average users is a distribution goal. You can worry for Joli getting lost among the other user friendly desktop distros because Joli actually going after that market space. I don't think anyone is loosing sleep over Backtrack not being in the top three visited distrowatch pages; it's not there purpose for developing the distribution. Altimately, what you provide is a false conflict. An erronious assumption that the purpose of anything which includes the Linux kernel must therefore be focused on retail market domination. Worse, that distributions who are actually focusing on a niche are failing for not being unsatisfied with the niche they focus on. Or that a given OS kernel is failing because some of the distributions that include it also happen to be focused on the desktop market. You just can't apply your uniform global economic motivations across all FOSSdome; those goals are not universal within the community. Being the gold cup holder of next year's F-1 curcuit is not the primary goal of all auto manufacturers and enthusiasts; it's not a universal goal across the entire auto industry. Heck, toothpaste brand XYZ is an utter failure given how minty brand ABC is. The fact that XYZ does not intend to be a mint flavored toothpaste is irrelevant. Really? This is the thought process your applying?

dcolbert
dcolbert

It is a Linux cop-out, smoke and mirrors. It seems disingenous and willfully deceptive. If you strip Linux from Debian, it is no longer Debian. If you strip Linux from Ubuntu, it is no longer Ubuntu. The kernel is the framework. Trying to draw a distinction between the two is just subterfuge. For practical purposes, as any Linux distro you want compares to Windows or Mac OS, from the perspective of an end user, the distinction between Windows, Mac OS and Debian is... Windows is Windows (is Microsoft), Mac is Mac (is Apple) and Debian is Linux (is Debian). Listen... the things an end user wants to do on Mac... it does it all, easy, well, reliably, to a level of satisfaction where most Mac users are satisfied enough to deal with the occasional headache. The same is true of Windows. So how come when we get to a Linux it always comes down to, "You're using the wrong *distribution* of Linux - you need this *specific* distribution... stop focusing on the Linux". That just makes no sense - which is part of why Linux can't catch on - on the desktop. And it is the biggest threat right now to JoliOS. JoliOS has the goods. It could be huge. But too many resources in the Linux community are focused on projects that compete with Joli OS in this space - and many of them are not good at all. MeeGo and Moblin and Netbook Remix and Easy Peasy and Puppy and about a dozen others all competing for mind-share, all overwhelming people looking for solutions. Meanwhile, iOS will be the Microsoft Windows of the next 20 years, and Android the Mac OS of this era... while Linux remains a novelty niche player fighting itself, cutting of its nose to spite its face. Which is too bad. Because JoliOS is an awesome OS platform, and Linux.

Neon Samurai
Neon Samurai

I don't know why you take such offense to the suggestion. If you where trying to pull a camper with a classic VW bug, I'd suggest that you may need to consider a different vehicle to toe your trailer. Why is it su unacceptable to point out that, like cars, distributions are built for different purposes. Put Red Hat Enterprise on a machine and claim it's not a good desktop OS; of course not, it's produced with the primary intent of being a server OS. Complain because Windows7 Home isn't doing all you need once installed on your server; well, sure.. maybe try a Windows distribution provided for use on servers. The difference between it and desktop focused distribution is a valid point of consideration. It's not about "your stupid for not using the right distro of the week". Your reading in some critisism when more applicable distributions are suggested; that's all you. If your complaining because your Pinto won't pull a two ton trailer; I'm going to suggest that you may need a more applicable vehicle. If your complaining because distro XYZ doesn't include proprietary drivers when it's clearly a Free Software focused distribution; I'm going to suggest that you may need a more applicable distribution. How is one supposed to handle that situation anyhow? Based on my understanding of the topic, the users could obviously be using a more applicable tool.. but heavens no.. don't point that out to them cause they'll just take it as a personal critisism and reject the whole topic area all together. That's been my whole point all along. Your attributing goals to an entire community which are not universal community goals and in most cases are actually completely oposing to community goals (freedom of choice). Your attributing your own interpretations and context to any information offered which may actually clarify your understanding of the community. When I say "that distro isn't really good for that because of XYZ, maybe try one of these which do focus on your needs" you are choosing to read "your doing it wrong. use this distro stupid." Look at this very example; based on your usual painting broadly with the same brush, Joli is an abismal failure because Canonical doesn't deliver stellar hardware support in it's Ubuntu distribution. And here I say; well, your focusing on the consumer level product not it's components and the distribution is the consumer level product. And the actual outcome? You had much grief with Ubuntu, you've had a great experience with Joli; at the distribution level, one can actually see the products and compare them on common attributes.

dford
dford

"it is so rough you have to write your own code just to get a directory listing." Yes - I did that that too! David

dcolbert
dcolbert

Linux has had opportunities as a pre-installed option several times. In the earliest cases, it was simply premature. Linux wasn't *ready* and this hurt the public perception of Linux in the long run. In more recent examples, Linux was just poorly executed. The Netbook example is my favorite example of something that was Linux's to lose - and that is exactly what Linux did - gave the Netbook market to Win XP and Win 7. If the earliest examples had come out of the box like JoliOS, things could have been radically different. But an Eee PC out of the box with the stock Linux OS did not have built in support for most popular codecs, and for most popular web-plug ins (like Flash) - and the package management was .rpm based and difficult - and most of the other Netbook Linux based OS platforms were more of the same. The only thing that JoliOS is missing is popular mainstream apps. They've addressed all of these other concerns that ended up getting Linux shown to the door on the earliest Netbook platforms. Standalone Facebook, Twitter. Angry Birds. If JoliOS could gain the support of the Intel AppUp market place - and the two of them could integrate tightly *and* get "big name" titles and app developers on board - that could be a major boost for Atom processors and Linux based mobile OS platforms. I think that is the most significant thing that is missing right now in this whole "space" - and that this is what makes or breaks JoliOS. I think it would make fantastic sense for both JoliOS and Intel AppUp and Atom processors.

dcolbert
dcolbert

Your Home or Windows key takes you back to the desktop from the current app while leaving the current app running. The Menu key (generally on the right side of the alt key to the RIGHT of the space bar) - brings up a pulldown Gnome menu that will give you access to the familiar "Accessories, Graphics, Internet, Office, Other, Sound and video, Wine, Places, System" menu. alt-F1 brings up a terminal. Ctrl-Arrow (left/right) switches desktops.

AnsuGisalas
AnsuGisalas

Long-standing "known issues" are a pain - but that, at least, is something we know and love from Windows, too. For me it's interesting that a linux distro has actually come as far as JoliOS, I mean, it's a pretty bleeding edge niche, and with no BIG NAME contender.

Neon Samurai
Neon Samurai

How is it willful deception? Smoke and mirrors.. a "cop-out" even.. really? So Debian is magically no longer Debian if I install it with the Hurd or BSD kernel instead of the Linux kernel? How does that little bit of political spin work now? Is Debian magically not Debian if it's kernel isn't running on an AMD or Intel processor too? Ubuntu would still be Ubuntu if they delivered it with a different kernel. The posix style system architecture and/or userland would provide the framework. The kernel is just a swappable comodity part; Canonical happened to choose the Linux kernel. It would still be Ubuntu if they'd chosen a BSD kernel in the beginning. It'd still be Ubuntu now if they swapped in a BSD kernel. If it's being specifically sold as a Linux based distro while not actually shipping on a Linux kernel then yeah.. that would definately fall under false advertising. I get that the public perception is that anything that happens to be based on a Linux kernel must therefore be the same product. This is a key reason I constantly suggest focusing on the distribution because the OS kernel is a commodity part in the assembly not the resulting product it happens to be included in. Why is the public perception not that the short comings of out brand of car are therefor common to all cars; maybe the Prius has a bad tail light often replaced in the first month after purchase, we don't then assume all cars ship with bad tail lights. Yet, in the OS world, it's very popular to make this same claim; lump all distinctly seporate distributions together under one category then claim one is representative of all. " So how come when we get to a Linux it always comes down to, "You're using the wrong *distribution* of Linux - you need this *specific* distribution... stop focusing on the Linux". " I'd say; because it's relevant. Because the consumer product is the distribution not what kernel it happens to use. Because different distributions are designed for different uses. Because it helps clarify options for system owners. Because the kernel is remarkably unimportant in comparison to the fully assembled distribution where user related decisions are actually made. Really? It makes no sense to consider that different distributions from different providers are different? How odd my thinking must be. In your own statement, you again suggest that all distributions have the same reasons for existance. "That just makes no sense - which is part of why Linux can't catch on - on the desktop. " - You suggest that the only reason to create a distribution or work on a given OS kernel is to "catch on - on the desktop". You dictate reasons why other's should or shouldn't be producing distributions. - And, it does make sense when you actually look at the distribution level. You can actually see that different distributions have different strengths. It's not actually the double-speak and spin you suggest it is. To be honest, you really shouldn't need to have any awareness of what kernel Joli is using. It's not relevant. If you pop in the disk, it loads and runs then who cares what kernel it's running? That's really the point. The item of interest is "Joli OS" not that it's a Linux kernel, not that it may be QT or GTK based GUI.

Neon Samurai
Neon Samurai

Actually, when compared at relevant levels, they seem about on average. Microsoft, as distribution provider, has two major distribution branches at current production version; Windows7 and Windows Server 2010. The Windows7 distribution is available in multiple "spins" or flavours; basic, home, pro, ultimate, media center. Two major current distribution versions and one broken down into five or so flavors. Apple has two current major distribution versions; osX and Ios. There isn't really any sub-categories | spins | flavours so the count is 2 distros of 1 flavour each. Debian; 6 Stable is the current production version. Canonical; Ubuntu Server and Ubuntu may qualify as two seporate distributions. Ubuntu, Kubuntu, Xubuntu would be "spins" or flavors of Ubuntu since they draw on the same repositories primarily differeing only by default UI. Counts seem to be pretty much on average across distribution providers. But this is the fundamental difference we always run into. I believe that you need to recognize the distribution level and look at that because that is where the consumer product exists and where a product provider has control over how the thing is assembled. It's not a bazzilion distributions put out by one over-extended organization. It's many different organizations putting out a distribution based on there development goals and target use. What's re-Donkulous is complaining about how there are two many Linux based distributions to choose from as one is walking out the door to choose one of twenty different coffee mixes from the same distributor. Starbucks just stopped servign the wonderous-maciato-extresso in Grande size; they really need to expand menu. Everyone wants more than three choices of internet service provider and preferably all three or more offering highspeed. Everybody happily accepts the various different subscription plans each ISP offers. Then everybody looses there mind over there being more than three choices of software platform available. This is why I agree fully with the fellow suggesting that distributions should drop "Linux" from the name more often. Far too much focus on an OS kernel rather than on products from seporate providers which happen to both make use of the same kernel name. It also results in the attribution of goals to all because those goals happen to be held by some.

dcolbert
dcolbert

Heh... I'm enjoying that Subject line in this thread. :) I don't think I had to try a different distribution. I had to wait for the Linux community to put together a distribution worth trying... Oooooh! Buuuuurnn..... :) Although, seriously... the problem is... When you compare... OS X has had, at best, two different flavors. OS X for desktop and for Server. Microsoft gets a LITTLE more varied... but basically you've got Win 7 and W2kX... You can get into details of Standard verses Enterprise, or Basic, Home, Starter, Pro, Ultimate... but those are all just trivial little degrees of difference designed mostly to justify charging more money. In general, you've got two different Microsoft platforms. A home version (XP/Vista/7) and a business server version (W2k8, W2k3). Linux is *reee-deeek-u-las*... Multiple different distributions with different platforms for different situations from single vendors, and lots of spin-offs taking a single distribution and "customizing" it more for one specific purpose. These outrageous websites and lists that turn into complex matrices explaining, "What Linux distro is right for you" is a great example of how the Linux distro model is applying ANYTHING but a "KISS" approach. Your car analogy breaks down. Ubuntu is a desktop OS. It should be able to do what Mac OS X or Win 7 does. And, meh... it has gotten way better at doing just that. But there isn't any compelling REASON to switch - especially when there are still some significant liabilities that remain when you DO switch (like, kissing streaming Netflix goodbye, if that is something you like to do). But... using your car analogy, they're selling it as a grocery getter.. and it is... but it is like a Kia... if you want Honda style on a budget, you're better off with a Hyundai. If you want Honda quality and aren't on a budget, buy the Honda. What model is left to make you decide to buy a Kia? When you're on a budget and want a car lacking in style? Linux comes up time and time again, in whatever "class" of OS you want to compare it to, as the brand marquee that offers LESS in your analogy. It is the thing you drive up in and go, "Well... it was the cheapest, and that was ALL I cared about"... and then get out and pop the headlight back into the fender only to watch the entire fender fall off. "I bought the Kia because I didn't want to be like all the rest of the posers driving around in Hondas, Hyundais, Nissans, Mitsubishis, and Toyotas". That is the OTHER reason, as I see it. You buy a Kia as a statement about your opinion of the competition - not because you think a Kia is a great car (well, and also because you're broke). I think this analogy fits Linux *way* better as a OS to Car analogy than yours. Heh. Now, part of what mixes it up with JoliOS is that this isn't something trying to compete DIRECTLY with Windows or OS X. Like I said, it competes more with Android and iOS. Comparing JoliOS to Windows is kind of a moot point. It doesn't actually equate well. But comparing to Android and iOS, it equates really well, and it doesn't do bad at all. It shows serious potential. In the bargain, it ALSO does things with Linux that would do Linux so much more good if they would adopt this on distributions that DO aim to compete with Windows and OS X. This is just another illustration that it COULD be done, but Linux idealists aren't willing to do the things necessary to make it happen. Their lofty, preachy ideals are hanging them up. Personally, I don't care. JoliOS is how Linux *should* work for end users, and they should continue to improve on the places where it still has some issues. It could be a fantastic platform.

dcolbert
dcolbert

But my primaly memories of the stories of the BBS guys I used to hang out with in the late 80s brought this concept to the surface. I know I've heard guys talk about having to do this, on very early 8 bit machines. Back when being a computer expert usually meant you were an electrical engineer by trade.

Editor's Picks