Ubuntu's death rattle? Please.

http://www.devside.net/blog/ubuntu-dying-2

I normally don't do this, but I think this is a good time to clear up some FUD this ignorant is spreading about.  You see, people like this are wolves dressed as sheep. They spread their ignorance and their lies about to influence you, thereby fulfilling their own prophecy.  So, I'm going to take the most obvious of his points and defeat them, because this is FUD, pure and simple.

1) Ubuntu has not turned a profit after 3 years, and probably never will.

This has the potential to be entirely true, but I theorize that Ubuntu was never destined to be a darling child and profitable, but Ubuntu is a platform-distribution from which others will be developed, and Ubuntu will survive through it's symbiotic relationship with it's children.  If this is Ubuntu's goal, then firm partnership deals may become a regular thing.  Companies will start paying Ubuntu because it is their platform, and platforms need to survive, either that or those companies will start dedicating time and resources to Ubuntu, to keep it afloat.  We have no fear.

Also, there are other options available for turning a profit in Free-software, and I have written a blog about this. LINK

Ubuntu may yet become a sponsored platform, a playing field which all of it's children have deemed important enough to keep alive,  I think that may have been Ubuntu's purpose all along.

2) Ubuntu's deal is a death toll

This is a sad pathetic excuse for a death toll if there ever was one.  Ubuntu is in that deal because it's what the people demanded of DELL. I think DELL has put as much energy into this deal as Canonical has, and both would like to turn a profit from it.  Hence the advertisements.   Canonical and DELL are just doing what we told them to do, and you're calling it a death toll?  FUD, pure and simple.  Ubuntu will settle into it's true purpose as time goes on, it will become a platform distribution, like Debian before it.

All in all, I found this article to be ridiculous, paranoid, and uninformed.  I hope that, in the future, people will think before they post such things...

Comparing Apple to Oranges... (Why Linux needs an Apple of it's very own)

I've many times written about what I think is going on in the Linux world, or what it needs.  Well now I'm going to write what I'm thinking again (like I'm supposed to do) and while not a highly new statement, I intend to show why Linux needs an Apple of it's very own.

First, you might ask why I'm bringing this up?  This is simple, Linux is growing, rapidly, but I don't feel it will be ANY of the major current players that truly bring it 'to the desktop'.  Let's look at the current recipe for desktop success, first.

Currently the two most successful operating systems known to man have these things in common:

Monolithic releases.
OS is a platform
An inclusive packaging format. (.dmg, .exe, etc.)
Integration.
Corporate backing.
Hardware vendor support.
Backwards compatibility

I'm going to wager that these things are what make for desktop success in terms of ease of use and adoption (I deliberately ignored the way that each OS vendor is making money on this).  Now, that I've spelled out what I think are important pieces of the puzzle, I'm going to explain why Linux as a desktop system, needs it's own version of Apple (metaphorically speaking)

So, I'll start with why I believe that the current developers of Linux are ignoring most of these formulas, although most distributions have the first one down, they don't do it right.

Monolithic Releases.

Most distributions that are popular for desktop users have a rapid, low-frills monolithic release schedule.  Often very little in the way of features added, but everything gets updated.  Releases need to be slower, more feature packed, and more exciting, in my opinion.  Maybe once every 18 months? Maybe a year? Maybe 3 years?  All I know is that slower is better when hardware vendors have to certify against each and every release.

OS as a platform. 

This is where the metaphor of Apple-for-linux is really apparent. Linux, as we know it, is not a platform. NO distribution has effectively BECOME a platform, and this needs to change. 

Current distributions are releasing what I would rather call a 'System application' than an 'Operating System'.  Why?  These systems BEHAVE more like applications than they do operating systems BECAUSE of this lack of being a platform.  Look at the most popular example.. Ubuntu. Ubuntu's repos are closed right from the very instant it releases.  Nothing new except bug fixes and updates.  This is the application behavior, in any application you will not see a major change until the next version releases.  So, what I see when I run Linux is a very capable, very full-featured application that is very modular. You simply add in applications from the repositories and you have extended the capability of the application.  A lot like you would see with a popular web browser... Firefox.

So, the question is, how can we turn this into a platform?  Well, firstly, an inclusive packaging format (read below) makes the age of the Operating System itself irrelevant.  Windows XP is 5 years old, but that doesn't really matter does it? No, because it's a platform upon which new software can be added, thanks in no small part to the inclusive packaging format.  Next is slower releases... you can't expect an entire ecosystem to form on something that's so rapidly changing that it's difficult to keep up with and certify against.  Slower, BIGGER/BETTER, releases are a way to help turn an operating system into a platform.  You need to let the dust settle and let people really get used to the new release, and then change things up after a couple of years...

An inclusive packaging format

Why does this help spell desktop success? Simple.  Currently, the only way to get software out for Ubuntu users, is to get it in the Ubuntu repos, and even then it's only available for people who use THAT version of Ubuntu and who's repos can satisfy the dependencies of your software.  This is problematic.   Each Linux project is typically understaffed in that regard, they spend TONS of time and energy putting together packages for their repos to try to satisfy user needs and continually fall short, they always miss something.

The burden of creating the software package for EVERY other operating system is NOT on the distributor of that system, it is on the developer of the application. Apple does not have to make the .dmg files for FLOCK but for some reason WE should have to make the .deb files for it? No, a standard needs to be available that is inclusive enough to make this trivial, much like .dmg for the mac.  The means that the a framework will have to be in place to support this new standard as well.

Now why would an inclusive packaging format be good for users?  Well, firstly is it makes the distributor, the distributor. Availability of the newest application is ALWAYS there, just go to the site, download the file, and you have it.  Next is that it makes most of the software dependency points moot, and unimportant, because it's inclusive.  It has most of the dependencies met, you run into the occasional error but it's usually not a huge deal.  Apple's .dmg is the status quo here.

The cons are that it's bloated of course... applications may load slower, will take up more space, etc.

Repos still have a place though, they're awesome for updating and maintaining a system, just not for installing applications or getting any exciting new programs.

Integration.

Actually, this is important to.  Novell's suse is an example of why integration is truly a great thing.  The whole system operates like a well oiled machine, rather than a bunch of spare parts forced to work together.  Applications are talking to each other, have a consistent look and feel, etc. This provides a much improved user experience.

Corporate backing

This comes in the form of marketing, management, and money.  Nothing special or new here, most products have this. \

Hardware vendor support

This is an important part of what 'The Linux Apple' will have to overcome.  One of the largest issues with Linux is flaky support from hardware vendors.  It's going to be this distributors duty to start making deals with DELL, HP, Nvidia, ATI, etc. to come up with a way to get hardware supported in this distro.  Perhaps DELL or HP offering customized versions of the distribution with support for their specific hardware? dunno.

Backwards compatibility

Simple, hardware and software I used before shouldn't stop working in the next few releases.
-----------------------------------------------------------

Now, it's on a new vendor to step up and offer this.  Ubuntu right now has the biggest mind share amongst the Linux distributions, but I suspect that Ubuntu would not be willing to do any of the things involving release schedules or packaging formats.  Red Hat won't do it, as they're off building their own sandbox to play in, and Novell, well, maybe they will do it.

The reason that I liken this future vendor to Apple is because they will be similar to apple in the sense that they will take something open source and work independently and make a lot of money on it, while the rest of the community continues off doing it's own thing... sort of like what Apple did with the BSD community.

Well, as always,

Rant Over, Flame On!

Adam.

The Shadowman says...

Greetings, faithful readers. Shadowman knows it's been a while since he's sent some love your way, but he's been busy. Places to go, people to see, things to do, yadda yadda. Shadowman won't bore you with the details. Here's what matters now: Shadowman is reporting live from the Red Hat Summit, bringing you the interesting topics of the day. Kind of like Edward R. Murrow, except without all the bombs and killing and stuff.

Got a question for Shadowman? Email him.

The entire world asks:

So Dell decided to ship Ubuntu on the desktop. Doesn't that bother you guys at all?

To which Shadowman responds:

Let's go back a few years, to the dawn of the Linux desktop.

Some of you may remember a guy named Havoc Pennington. Those of you who can remember the murky past of the Linux desktop will recall that Havoc was one of the founders of the GNOME Foundation. He was also its first chairman. He also founded freedesktop.org, to help bring together the GNOME and KDE communities through the shared and comforting love of open desktop standards. He was kind of a rock star.

It's not like Havoc went anywhere. It's not like he retired from the game, like some sort of geek Sandy Koufax in his prime. He's still at Red Hat. Which might make one wonder: if Havoc Pennington is still at Red Hat, then why isn't Red Hat like, totally owning the desktop?

Funny thing about experience, and how it shapes your beliefs. Nobody spent more time trying to bridge the divide between Linux and Windows than Havoc Pennington did. And after all those years, what did Havoc conclude?

His widely circulated quote on the matter: "It's easier to win on user value than it is to match technical castles of unbounded size." You can Google it.

So what does this quizzical expression mean? To Shadowman, it means a couple of things.

First, it's the articulation of a view, gained through years of experience, that matching the incumbent feature-for-feature is inherently a losing proposition — especially when there are a number of areas where Linux will never, ever completely catch up. How many of you, gentle readers, still dual-boot Windows systems to play games? And has that changed substantially from five years ago? Ten years ago?

Second, it's a compelling argument that the real way to win the battle for "the desktop" is to change the battleground itself, and not necessarily to a ground where current Linux users will expect. The best desktop minds at Red Hat are now focused on pursuing Havoc's line of inquiry: "win on user value".

Where is "user value" now? Online, that's where. And moreso with every passing day.

Example number one: One Laptop Per Child. Example number two: Mugshot. The goal of each: to fundamentally remake the "desktop" experience to exploit the "always connected" reality of computing today.

Crazy ambitious? Yep. But you should have heard Brian Stevens talk about that ambition this morning, and how single-mindedly Red Hat engineers are pursuing it.

Chris Blizzard and Havoc Pennington both did their time as champions of Ye Olde Linux Desktop. Now they are leading the way to the New Online Desktop: Mr. Blizzard in OLPC-land, and Mr. Pennington in Mugshot-land. And if this means that Ubuntu claims a perceived high ground in Ye Olde Linux Desktop — at least for now — then so be it.

To many, this sounds like surrender. It's not. Shadowman is continually amazed by the people who assume that Red Hat just doesn't care about the consumer desktop, and how easily people assume that Red Hat doesn't have the chops to compete for the consumer desktop, and how Red Hat isn't smart enough to see Ubuntu coming.

When Shadowman hears about the Dell survey that (allegedly) led to Dell's decision to preload Ubuntu, he can't help but think of the old quote from Henry Ford: "If I'd asked people what they wanted, they would have asked for a better horse."

Microsoft is not afraid of Ubuntu, folks. They are afraid of the pretty green toy that will teach kids a fundamentally new way of using technology. They are afraid of online services, and the explosion of collaboration that those services will enable.

Look at the big Microsoft stories in the news recently. Big story number one: Windows licenses for $3 in the developing world, trying to counter OLPC. Big story number two: the potential purchase of Yahoo, trying to counter Google and their gigantic lead in online services.

So let Ubuntu take the lead in building the better horse. They've earned that lead, and good on them. Meanwhile, at Red Hat, the desktop engineers are working day and night to invent the automobile. And the first model T is almost ready to roll off the line.

Ubuntu fanboys: flame away. Shadowman loves you all regardless. And you know this, man.

This appeared in my RSS-feed reader yesterday morning, and I just want to say that I think the idea is right.  As our lives slowly move more and more online, and connectivity becomes more and more of a given than an option, our operating systems should adapt to this.  I would actually not mind my desktop playing nicely with Google Apps, Facebook, and the other multitude of online applications I use.

These kinds of things ARE important, however, is Ubuntu really building 'the better horse' here?  Let's not forget that Red Hat is (beyond any shadow of a doubt) an open company, and because of the very nature of open-source they will never be able to maintain a technical advantage over their competitors for very long.  Ubuntu as an established brand could quickly adapt Red Hat technology and create the 'Ubuntu Global Desktop'. 

I am curious what they're going to do for this in the future, we'll see...

How can we exist in a world without DRM?

     Interesting things I'm reading happening with DRM lately, like the discovery of the HD-DVD encryption key.  Well, obviously this whole DRM thing is a HUGE waste of money, it's just not working the way it should.  So the question I'm going to pose here is, how can we exist without DRM?  How can we live in a world where software is simply given away?  Well, I think there are a few ways that applications can be given away, and with free source code as well.

Subscription Services

This is the biggy right now.  Subscription services, support, etc. are a big business for businesses.  Home users may not care as much about support services, but perhaps other services can be devised that cater to needs of home users as well. Perhaps with a photo application you would like to have access to a slab of web-space to upload your photos to? Maybe something that you can share with friends and family, and update from.  It would be nice to see that slide show I have of my family and friends updating itself.  The options are there to get creative with services.  What are the merits to a service based model? Well, you're the only throat to choke for that service, really.  No two companies have exactly the same support service, and this is the same deal.  Your competitors can take your source code and do whatever with it, but really they wont' be able to compete as well without the ability to create a service like yours.

Advertising

Ok, this one is scary to people, and a little harder to implement.  There are several ways you can 'advertise' with a program, and it can be really lucrative without being invasive. Let me see if I can come up with an example that isn't sure to cause paranoia.  Let's say that GAI's popularity contest data is showing that you're installing (and LOVING) a LOT of games, perhaps a suggested applications section will point you to the latest and greatest Linux compatible MMORPG? It's possible.  Also, perhaps there are subtle adverts hidden inside of a game?  Coke-a-Cola machines that you can interact with, billboards, etc.  Ok, and on the operating system front... there is crap-ware.  On a freely distributed operating system, Crap-ware would probably be a lucrative deal.  Maybe AOL wants to get more Ubuntu-users using Roadrunner, etc.  There is a lot of room here for give and take, some adverts are too much, and others are just right.  If any distribution or application goes this route, tread carefully.

SALES!
OK, OK, I know what you're thinking. "You can't SELL open source software! That JUST doesn't work!".  Wrong.  Sales in applications are all going to be about branding, I'll give you an example:  Go into your Breakfast cereal aisle at your local grocery store, you'll see the boxes of cereal, and then the bags of cereal.  The bags contain SO much more cereal than the boxes do, they're cheaper, and they taste the same (many times they're exactly the same cereal).  But yet the 'budget' cereal isn't purchased nearly as often as the boxed version, why? Branding and marketing.  If I create an open-source photoshop that is FREELY distributable, GPL, etc. and I build that into a strong brand, I bet that even though it's open source my competitors taking my code wouldn't hurt me much, through the establishment of a strong brand.


I'm willing to bet that a combination of the three would be extremely lucrative, too.  The great thing about software is that there exists the option to meet any single demand amount, so you have to find a way to take advantage of that mass distribution without upsetting the users to much that they create an alternative.   Well, that's my slightly ignorant brain-storm for the moment, as always:

Rant over, Flame on

Adam.

Sorry for the absence, here's a recap :)

So, it's been a while.  I didn't mean to to be gone so long.  So, what's happened since my last blog?  Well, for starters the DELL deal with Ubuntu.  Great news right?  Eh, probably.  I'm kind of ho-hum about it myself, and I'll remain that way until I see a reason to think otherwise.

While I'm terribly excited about Ubuntu being offered, I just can't see DELL following through with the execution.  I expect a half-assed attempt to sell these boxes, and when there is an inevitable failure thanks to the fact that they're carefully hidden, DELL will declare Linux 'not ready for the desktop' officially.  So, here's hoping that DELL doesn't screw this up!  Hopefully I'll be wrong :)

Should your OS be trivial? Part 2

Sorry for the amount of time it took in between posts, life as usual I suppose, haha.  So, to re-cap in my last post I mentioned how I felt that Window's platform approach was better (in some ways) for the application enthusiast, and how the non-triviality of Linux (Ubuntu by example) is not friendly for the person who enjoys playing around.  Well in today's blog I want to discuss how the triviality of Windows makes the switch to Linux difficult and may result in the problems we are commonly hearing about.   I know when it comes to these issues, it's hard to not spew forth redundant complaints and information- but I'm going to try to avoid that.

Now, to the question:  Why is Windows triviality causing our common complaints.  Again, it's an invisible system, it doesn't matter.  No one CARES about Windows they just use it. When it comes right down to it the law of the land is if you want something done either do it yourself or pay someone.  Well, people are used to the latter, and not the former.  In other words, people would rather shell out cash and have it done for them.  Not ALL people, but a majority seem to prefer this approach.  I mean seriously, how many birthday presents have you gotten with the words "some assembly required" on them, and you never used them until someone else put it together for you.

Well, Linux is that present.  It's "some assembly required", and a lot of people would be MORE Than happy with it after the assembly is completed, it's just getting through that assembly that bothers people.  So people come in "Why doesn't mp3 just work!", this that and the other, it's not because mp3 is missing. It's not because FLASH is missing, it's because there is assembly required and a lot of people just don't want that.  Despite the legal reasons behind these actions.  These people are not used to being involved in a community, they're used to buying from a corporation.  They are used to their operating system not being something to care about, in other words, if the option existed to eradicate the operating system and have programs that "just worked" I bet they would do it.  These people don't care.

You (the reader) and myself, are probably in another group that do care.  We're the tech enthusiasts that care about our operating system and want to know about it, and Linux is perfect in that regard.  And when you give someone the gift of Linux you need to add the disclaimer "some assembly required".  To further the analogy:

• Windows is the toy with no assembly required.  It isn't as much fun as another toy with thousands of pieces and options, but it's a lot easier to get started playing with.

• Gnu/Linux is the toy with assembly required.  It isn't as much fun to put it together, but once you do you're going to have a blast with it.  Step 1 is the problem.

So again, I'm going to argue that people aren't missing their mp3 collections, or their flash, or their DVD playing.  They're fighting with the fact that the OS isn't trivial like windows.  In GNU/Linux your OS needs to be watched, and maintained, and put together, and used, and then replaced in 6 months to repeat the process.  So the question now is, if make it trivial for them, does it matter anymore?  Look at Linspire, that's a perfect example of a trival Linux distribution, and the ONLY reason they're in the news is when they do something that is distasteful, etc.  I'm going to propose that becoming more trivial for the user only causes problems in the end, and that the way GNU/Linux is headed right now is more suited to the way things should be.

I suppose there will always be a market for trivial, the question is though, can a solid trivial competitor to Linspire be created?  Can a respectable company who treats their clients right step up and make a trivial version of Linux that doesn't make you want to shoot yourself after 10 minutes of use?  Well, perhaps the people at Linux Mint need to start working on a business plan, because they're about as good a competitor as you can get right now.

I hope that a company like Red Hat, Novell, or Canonical would be willing to help produce a trivial OS.  Since those are all respectable companies, but we'll see.

Should your OS be trivial? (part 1)

What is it about Windows that makes people tolerate the 5 year release cycles? This thought has been on my mind for a while, and I'm hoping I can spell out some issues that I think a lot of new users of GNU/Linux are experiencing. I think the complaints we hear (mp3, DVD, etc.) are not themselves the problem, but there is a deeper underlying issue. I believe it's this same issue that causes distro-hopping behavior in Linux, etc. So, hopefully I'll be able to make you can see the point I've just come to.

Windows is trivial. There, I said it. The operating system just simply does not matter, and new users of GNU/Linux are simply not used to having an operating system that matters. They are not used to living in a world with 6 month release cycles, frozen repos, etc. They are used to the operating system essentially being ignorable until something breaks. At the point where something breaks they'll call a friend and have it fixed, back to business as usual. The operating system stays out of their way, and they like it that way, even a lot of tech enthusiasts like this system.

Now, how is Windows to blame for distro-hopping behavior? Well, a tech enthusiast coming from windows didn't have the opportunity to switch operating systems every 3 days. But they DID have the opportunity to try out TONS of new software without any major effects on their system (usually). As a Windows user I had to have the latest IM client, the latest beta version of FireFox, and the IE7 beta. A windows user can try out new software ALL the time, because windows is trivial, it doesn't matter.

We have been trained to perform a set of actions, when we use Windows we're not using the OS we're simply using the applications. Window's bare-bones install left a lot to be desired, but if you were an enthusiast who didn't use GNU/Linux what you DID do was dream about using an iMac while you downloaded the latest beta you discovered. For me. It seems like every other week that I had completely replaced every single software application I used with something newer and more shiny, and resource hungry.

Well, as a windows user for, well, my entire life. This behavior is ingrained in my sub-conscious, and I'm not the only one that this happens to. A quick glance in the Ubuntu forums shows me that a lot of people spend their spare time playing with other distributions. What for? Because it's the same thing they used to do on Windows. When thrown in this situation with frozen repositories (for good reasons, however) you get a user who feels kind of caged by the system. "Life doesn't exist outside of these repos" and so they look onward for another system to play with, something more 'free', something less 'halted in time'.
When the system isn't getting anything new, it's just not fun anymore.

In the Linux world when a distribution is released, that's it. No more exciting software, you have the thousands of packages in the repos, and any mysterious combination of them COULD break your system, or cause ill behavior. Each distribution becomes a frozen platform, stopped and ready to use. Ubuntu as an example is frozen, save for bug fixes and security updates. Am I saying this is a bad model? absolutely not. Now, you say "why don't you compile from source?", simply, that's 'non-trivial'. It's a foreign method of installation that may or may not play nicely with the package management in your distro, and using newer dependencies could break something very easily. The system for a tech enthusiast is no longer a platform like windows was, it's more like an application. Ubuntu seems less an OS when it's frozen like that, and more like a very very very nice application. But you won't see any changes until the next version, and if you're like me, you'll be running the beta as soon as it's usable.

Yes the repositories include tons of stuff to play with, but in the end, it still feels a little boxed in. You know that for next 6 months, two years, or until the next release, you're not going to see anything new or exciting. Perhaps that's what users like myself are looking for, we're looking for a distribution that isn't just a great application, but an operating system, a platform. We want to see new software coming down the pipe every so often, just something new to play with.

Well this blog has gone on long enough, prepare for part II coming to an unread blog near you! In part two I'm going to detail how I feel this same "platform to application" transition is causing a lot of the complaints that are commonly dealt with in many 'desktop' distros. As always,

Rant over, flame on.

Adam.