Ensuring the Success of Dell Desktop Linux

[karellen]

http://itmanagement.earthweb.com/netsys/article.php/3672516
this guy seems pretty realistic and well-intended to me...
what do you think?
(I really want Dell to have succes in their task...Linux installed on their boxes...you know all the story...)

[needtolookatascreenshot]

Well, I think the words well-intended and reallistic and Robert Pretenderlde can't be used in the same sentence.

http://www.newsforge.com/article.pl?sid=04/08/11/180256
:lolflag: :shock:

[karellen]

Well, I think the words well-intended and reallistic and Robert Pretenderlde can't be used in the same sentence.

http://www.newsforge.com/article.pl?sid=04/08/11/180256
:lolflag: :shock:

:confused: ....ups...I didn't know about that...:D

[mac.ryan]

Interesting article. He has a point, yet I believe he missed a few things.

Dell is not a nice mummy trying to fulfill the wishes of her children for Christmas eve... Dell is the #1 computer vendor in terms of numbers of computers an possibly gross revenue. If they are exploring a "linux line of thinking" it is because they see there a chance to make more profit, not because they want to fulfill the wishes of a group of users that is probably below 1% of their potential customers. The article seems to suggest that potential linux Dell buyers should be "kind with Dell"... that's stupid: Dell is a multi-billionaire company: they should come up with a good solution or take the blame for a bad one.

GNU/Linux users do not need a Dell distro. They just need to be reasonably sure any distro will work on a Dell machine. This can be achieved easily by Dell: the selling volume of Dell is high enough to lobby effectively on hardware producers for the release of open drivers. Unsupported hardware (or hardware configuration) is probably the #1 problem in getting a GNU/Linux system going...

GNU/Linux is not a different OS, it is a different approach to OSs.
Ultimately, Dell's bigger mistake is trying to replicate the Windows "slave" mentality on GNU/Linux. GNU has been invented because people wanted more freedom, not because people wanted cheaper software. If Dell will try to lock up users in a "Dell Linux" distro/support/community mentality, then I can already see they will fail. If they will aim to guarantee more freedom to their users, they will succeed.

In a nutshell... Dell best option would be IMO to invest on lobbying for (or development of) open drivers for their hardware + developing a few open packages for managing their specific features (i.e. the DirectMedia key). The GNU/Linux community better option is to keep on thinking critically.

[Tundro Walker]

I can't find the article, but Joel on Software...

http://www.joelonsoftware.com/Archive.html

...posted an article talking about what it takes for market penetration. Basically, you outline what's keeping people from adopting your product. When there's already an entrenched product, in this case Dell pre-packaging MS Windows, there are huge hurdles to to leap over. One of those hurdles is getting Dell to think about pre-packaging Linux, which it looks like they're planning on experimenting with.

But, another hurdle is a whole slew of MS Windows users locked into that proprietary format. Are they just going to give up all the work they did in MS Word using word.doc format, or spreadsheets with .xls format? What about music they worked on using whatever proprietary software that only works in Windows?

This puts double-work on Ubuntu, and Linux distro's / open-source software in general, because they not only have to focus on creating their own world that's good, but making it compatible with the pre-existing, dominant force, making migration easier and more portable. With things like OpenOffice, it makes the migration less painful, since it can read some of MS Office's proprietary format.

I think 2 factors will help push Linux along into the spotlight as an OS to contend with.

1) Web-apps are more and more dominant, because they provide true portability and cross-platform compatibiity.

2) Windows continued poor treatment of customers, and continual increase in cost to upgrade will keep forcing some customers away.

3) The continued propogation of small, powerful, multi-function "accesory" devices...mp3 players that also play music and have calendar control, cellphones that play music and calendar and surf the web, game consoles for gaming, etc.

Desktop computers are a bit like the Industrial Revolution, in that there was a great migration of stuff to a single desktop system that did everything. It was the only thing powerful enough to keep up with continued advances. But now, you can carry powerful devices with you that do most of what you want, and dock them with a desktop for transfer and sync. In essence, our desktops are like downtown, and more and more folks are opting to spend more time in the suburbs by using the cellphones, mp3 players, palm-tops, etc for more of the utilitarian functions.

Even if a person only uses a desktop, a 2ghz CPU with 512mb or 1gb of ram is pretty much all the computer power an "average" user would need...average meaning "not a gamer" and "not a 3d graphics developer". Average user is one like grandma, who just wants a simple machine for email, word processing, web surfing, music, maybe spreadsheets, etc...utilitarian things.

It's difficult to justify a more powerful machine, but Microsoft can, since each upgrade has more bells and whistles on it that are usually just eye candy, or them reinventing the wheel in a more complex fashion to justify more powerful hard-ware, and the need to buy more powerful anti-virus, spyware, etc packages that use up a lot of overhead to protect Windows from getting infected or hacked.

When you take away the huge overhead, you have more than enough system resources on a 2ghz machine to do whatever you want...if you're running something like Linux. Linux supports plenty of old and new hardware, and distro's like Ubuntu continue making an overall OS / Application roll-out experience more user friendly, updated with newer innovations, free of bugs, etc while still working on your current hardware.

So, the justification to buy bigger, better machines is tapering off, especially when folks actually look at the cost comparison. Let's see, I can keep my current computer, and either run Ubuntu for free, or use something like Xandros for ~$50, both of which are functional, modern OS' that come packed with lots of free software to cover my uses. Or, I can buy a new $1500 computer, and $300 for Vista, and plop another continual $1000 over time to either maintain current licenses for software I use, or to update them as they figure out they can charge folks again for "Vista" support or a "Vista Compatible" version, which is basically like the old version, expect designed to be hobbled if you keep using it on an older Windows OS version.

People will vote with their pocket books. But, a large hindrance is a lack of information...well, a lack of the average user, who just wants a computer that works and doesn't want to spend any more time in the matter than is absolutely necessary to research such. Linux distro's can't exactly foot the huge marketing bill that Microsoft pays every day to stay in the spotlight.

Seriously, USB cables at CompUSA say "Vista Compatible!"...they're USB cables...that's like saying a certain brand of gas is "Honda Compatible"...and charging folks more for it. Linux doesn't have that kind of penetration.

But, if Microsoft keeps doing things in their own best interest rather than the users, keeps over-charging for software, keeps treating their customers like crap...folks will revolt. It will be gradual, and they will shift to Apple, or Linux, but as it happens, Microsoft will wake up and realize they need to do something different. Currently, they're strategy is vendor lock-in, by getting folks locked into their proprietary format. But, they see folks moving to Linux, so, not being dumb, they collaborate with Novell to come up with a Linux solution. You want Linux? Fine, you can have it...under our terms...(which will probably end up either putting a bad mark on the Linux name if they advertise a version like it's really great and end up making it really bad, setting Linux promotion back 5 years, or it'll basically be Microsoft Windows/Linux, blurring the line between the two, and basically bringing Microsoft's worst practices...proprietary formats, vendor lockin, non-portability/compatibiltiy, non-scalability, etc, over to their brand of Linux, which will get the biggest spotlight since they have the marketing funds to promote it.)

It's hard to tell what's going to happen. But, personally, I don't think Dell will do well with their Linux endeavor, because Linux isn't really a "new desktop" thing. Linux is more about making sure what you have still works good.

I think a company that buys up old hardware and refurbishes it by doing some "21 point check" on it to make sure it's all still good, then installs a Linux distro on it and selling the boxes for $400-500 will start making a killing. Sort of like the "1 laptop per child" charity, except a business that offers very cheap, refurbished computers...with Linux installed.

And, if a company, like Geek Squad, would just start making Linux support calls, that would cover user worries about where they'd turn to for support. In fact, the refurbish company could contract with someone like Geed Squad to provide tech support for their Linux builds.

Another place where I think Linux will make more of a show will be embedded devices. Linux is already used in a lot of these, but what I'm talking about is people who just want to go buy a generic mp3 player or cellphone, and install their own OS and applications on it like they do a normal computer. Folks are tired of their crappy firmware on their mp3 players, so they use things like RockBox to get it to do what THEY want it to do. I think more Linux distros that come out focusing on allowig users to install it on their embedded device for tweaking would increase more light on Linux.

More focus on thin-client or wifi-client would also bring more Linux to light. Another one of that guys articles talked about 7 applications to make life easier for business to transition to Linux. One talked about a thin-client program letting 1 main computer fuel up to 10 thin-client desktops. Sounds great. But, thin-clients didn't catch on in the past, because it was pretty cheap to just buy a real PC and install the whole software on it for each employee's use. But, as Windows keeps screwing up with their customer support and lack of portability / cross-platform compatibility, more companies turn to web-based apps. And mobiles execs on the go like to use palm tops and cellphones for their calendars and such. In trying to re-plant the thin-client model back into mainstream business, you can skip right over it and simply go right to wifi-thin-client. Give users a wireless thin-client that connects to a main hub.

In other words, Linux has always had heavy penetration in server & embedded device use. As more businesses (and homes even) use more portable embedded devices for work, we'll see Linux make an influx. And as more wifi-client comes about, Linux will be there to spear-head. With that, I don't think it's so important whether it can or can't compete with Windows in the desktop world, because Microsoft is struggling to compete with it in the server and embedded device world. Linux will always be around, and maybe it will or won't make more head-way into desktop computing. Personally, if they keep making LiveCD's so easy to install, I could care less if some computer company pre-installs OS on the machine...the first thing I'm going to do is install the OS I want using a LiveCD. Now, if they charge me MORE to install an OS, I'll say "no thanks, just give me the hardware", and I'll do the rest.

[Tundro Walker]

Huh, I just realized...Microsoft already leverages Dell to pre-install Windows. So, since Microsoft now wants to compete with Linux by using Novell, Dell will most likely only install Novell Linux as requested, which, as I stated in my previous response, will probably end up being some kind of bastardized hybrid Win-nux thing that might end up giving Linux a bad name. (Which, for a long-term campaign to mess up the name of Linux so folks won't want to use it in favor of Windows for the next 5 years may be worth the cost for Microsoft...never know. I wouldn't agree with that kind of "unethical" practice, IE: producing a Linux OS that sucks so everyone things Linux sucks and runs back to your main OS..Windows. But, Microsoft is no longer a business that lives paycheck to paycheck. They're an investement firm. They can afford to invest in a failed side project if it means greater long-term income somewhere else. Wow, this sounds kinda Illuminati'ish...

[karellen]

I can't find the article, but Joel on Software...

http://www.joelonsoftware.com/Archive.html

...posted an article talking about what it takes for market penetration. Basically, you outline what's keeping people from adopting your product. When there's already an entrenched product, in this case Dell pre-packaging MS Windows, there are huge hurdles to to leap over. One of those hurdles is getting Dell to think about pre-packaging Linux, which it looks like they're planning on experimenting with.

But, another hurdle is a whole slew of MS Windows users locked into that proprietary format. Are they just going to give up all the work they did in MS Word using word.doc format, or spreadsheets with .xls format? What about music they worked on using whatever proprietary software that only works in Windows?

This puts double-work on Ubuntu, and Linux distro's / open-source software in general, because they not only have to focus on creating their own world that's good, but making it compatible with the pre-existing, dominant force, making migration easier and more portable. With things like OpenOffice, it makes the migration less painful, since it can read some of MS Office's proprietary format.

I think 2 factors will help push Linux along into the spotlight as an OS to contend with.

1) Web-apps are more and more dominant, because they provide true portability and cross-platform compatibiity.

2) Windows continued poor treatment of customers, and continual increase in cost to upgrade will keep forcing some customers away.

3) The continued propogation of small, powerful, multi-function "accesory" devices...mp3 players that also play music and have calendar control, cellphones that play music and calendar and surf the web, game consoles for gaming, etc.

Desktop computers are a bit like the Industrial Revolution, in that there was a great migration of stuff to a single desktop system that did everything. It was the only thing powerful enough to keep up with continued advances. But now, you can carry powerful devices with you that do most of what you want, and dock them with a desktop for transfer and sync. In essence, our desktops are like downtown, and more and more folks are opting to spend more time in the suburbs by using the cellphones, mp3 players, palm-tops, etc for more of the utilitarian functions.

Even if a person only uses a desktop, a 2ghz CPU with 512mb or 1gb of ram is pretty much all the computer power an "average" user would need...average meaning "not a gamer" and "not a 3d graphics developer". Average user is one like grandma, who just wants a simple machine for email, word processing, web surfing, music, maybe spreadsheets, etc...utilitarian things.

It's difficult to justify a more powerful machine, but Microsoft can, since each upgrade has more bells and whistles on it that are usually just eye candy, or them reinventing the wheel in a more complex fashion to justify more powerful hard-ware, and the need to buy more powerful anti-virus, spyware, etc packages that use up a lot of overhead to protect Windows from getting infected or hacked.

When you take away the huge overhead, you have more than enough system resources on a 2ghz machine to do whatever you want...if you're running something like Linux. Linux supports plenty of old and new hardware, and distro's like Ubuntu continue making an overall OS / Application roll-out experience more user friendly, updated with newer innovations, free of bugs, etc while still working on your current hardware.

So, the justification to buy bigger, better machines is tapering off, especially when folks actually look at the cost comparison. Let's see, I can keep my current computer, and either run Ubuntu for free, or use something like Xandros for ~$50, both of which are functional, modern OS' that come packed with lots of free software to cover my uses. Or, I can buy a new $1500 computer, and $300 for Vista, and plop another continual $1000 over time to either maintain current licenses for software I use, or to update them as they figure out they can charge folks again for "Vista" support or a "Vista Compatible" version, which is basically like the old version, expect designed to be hobbled if you keep using it on an older Windows OS version.

People will vote with their pocket books. But, a large hindrance is a lack of information...well, a lack of the average user, who just wants a computer that works and doesn't want to spend any more time in the matter than is absolutely necessary to research such. Linux distro's can't exactly foot the huge marketing bill that Microsoft pays every day to stay in the spotlight.

Seriously, USB cables at CompUSA say "Vista Compatible!"...they're USB cables...that's like saying a certain brand of gas is "Honda Compatible"...and charging folks more for it. Linux doesn't have that kind of penetration.

But, if Microsoft keeps doing things in their own best interest rather than the users, keeps over-charging for software, keeps treating their customers like crap...folks will revolt. It will be gradual, and they will shift to Apple, or Linux, but as it happens, Microsoft will wake up and realize they need to do something different. Currently, they're strategy is vendor lock-in, by getting folks locked into their proprietary format. But, they see folks moving to Linux, so, not being dumb, they collaborate with Novell to come up with a Linux solution. You want Linux? Fine, you can have it...under our terms...(which will probably end up either putting a bad mark on the Linux name if they advertise a version like it's really great and end up making it really bad, setting Linux promotion back 5 years, or it'll basically be Microsoft Windows/Linux, blurring the line between the two, and basically bringing Microsoft's worst practices...proprietary formats, vendor lockin, non-portability/compatibiltiy, non-scalability, etc, over to their brand of Linux, which will get the biggest spotlight since they have the marketing funds to promote it.)

It's hard to tell what's going to happen. But, personally, I don't think Dell will do well with their Linux endeavor, because Linux isn't really a "new desktop" thing. Linux is more about making sure what you have still works good.

I think a company that buys up old hardware and refurbishes it by doing some "21 point check" on it to make sure it's all still good, then installs a Linux distro on it and selling the boxes for $400-500 will start making a killing. Sort of like the "1 laptop per child" charity, except a business that offers very cheap, refurbished computers...with Linux installed.

And, if a company, like Geek Squad, would just start making Linux support calls, that would cover user worries about where they'd turn to for support. In fact, the refurbish company could contract with someone like Geed Squad to provide tech support for their Linux builds.

Another place where I think Linux will make more of a show will be embedded devices. Linux is already used in a lot of these, but what I'm talking about is people who just want to go buy a generic mp3 player or cellphone, and install their own OS and applications on it like they do a normal computer. Folks are tired of their crappy firmware on their mp3 players, so they use things like RockBox to get it to do what THEY want it to do. I think more Linux distros that come out focusing on allowig users to install it on their embedded device for tweaking would increase more light on Linux.

More focus on thin-client or wifi-client would also bring more Linux to light. Another one of that guys articles talked about 7 applications to make life easier for business to transition to Linux. One talked about a thin-client program letting 1 main computer fuel up to 10 thin-client desktops. Sounds great. But, thin-clients didn't catch on in the past, because it was pretty cheap to just buy a real PC and install the whole software on it for each employee's use. But, as Windows keeps screwing up with their customer support and lack of portability / cross-platform compatibility, more companies turn to web-based apps. And mobiles execs on the go like to use palm tops and cellphones for their calendars and such. In trying to re-plant the thin-client model back into mainstream business, you can skip right over it and simply go right to wifi-thin-client. Give users a wireless thin-client that connects to a main hub.

In other words, Linux has always had heavy penetration in server & embedded device use. As more businesses (and homes even) use more portable embedded devices for work, we'll see Linux make an influx. And as more wifi-client comes about, Linux will be there to spear-head. With that, I don't think it's so important whether it can or can't compete with Windows in the desktop world, because Microsoft is struggling to compete with it in the server and embedded device world. Linux will always be around, and maybe it will or won't make more head-way into desktop computing. Personally, if they keep making LiveCD's so easy to install, I could care less if some computer company pre-installs OS on the machine...the first thing I'm going to do is install the OS I want using a LiveCD. Now, if they charge me MORE to install an OS, I'll say "no thanks, just give me the hardware", and I'll do the rest.

one of the best posts that I have read for the last weeks :)

[prizrak]

If they can sell servers with Linux to businesses and make money along with providing support they can do the same with desktops. Simple as that.

[mac.ryan]

Seriously, USB cables at CompUSA say "Vista Compatible!"...they're USB cables...that's like saying a certain brand of gas is "Honda Compatible"

Unluckily it is not. There is a compatibility standard for cables in Vista: because of the tilt bits stuff, a cable that doesn't guarantee certain electric criteria (e.g. electric dispersion not superior to X) will trigger a reaction to the system as if somebody would be putting a voltmeter on the circuits... so the system will stop working. It is sick, I know, but this is where M$ is at in its attempt to seduce the majors of cinema industry.

Apart from this small technical remark, I liked you post, although I disagree from the basic approach of marketing GNU/Linux as a commercial product. GNU/Linux is grounded on a cultural/social/political vision, and IMO advocacy of its use should be based on those principles.

The biggest strength of any FLOSS project is in its freedom. If you get to the point of gauging it with monetary/economical parameters you will automatically diminish its freedom, as you will link to criteria of economical productivity/efficiency rather than to those of effectiveness and - eventually - elegance.

(Of course, I am not saying that the economical dimension of FLOSS should be overlooked... just it shouldn't be central in it's marketing)

[prizrak]

although I disagree from the basic approach of marketing GNU/Linux as a commercial product. GNU/Linux is grounded on a cultural/social/political vision, and IMO advocacy of its use should be based on those principles.

If you can convince a company to use Linux based on those principals you are the best salesman ever. Dell isn't planning on selling Inspirons with Linux, it is planning on selling Latitudes. Reason for it is that a home user won't switch to Linux unless she is a geek and lets face it geeks don't need preinstalled Dell systems. We know of System76 and how to find compatible hardware. Desktop adoption will be spearheaded by corporations (same way as the PC was). Corporations don't give a tiny rat's behind about political and philosophical roots and ideals of GNU they care about ROI. If the ROI of rolling out a Linux system is greater than a Windows system then they will switch. If it is not, then they won't.

[mac.ryan]

Corporations don't give a tiny rat's behind about political and philosophical roots and ideals of GNU they care about ROI. If the ROI of rolling out a Linux system is greater than a Windows system then they will switch. If it is not, then they won't.I agree with you, although the point is not really about the ROI on "rolling out" a system (if I understand correctly this expression), but about the ROI in the swap between the two (which is a cost in either way you perform it), given that the TCO is however lower on GNU/Linux than on Windows.

Exactly for the reason you explained, however, I am convinced that there is no much point in doing marketing for Linux with a company... if they are well managed, they will go for the better option! In fact: GNU/Linux with basically no marketing is expanding, MS Windows (that poured 500 U$ millions in vista marketing only) is not.


If you can convince a company to use Linux based on those principals you are the best salesman ever.

I did it twice, already! :) [however, to be honest, they were both companies working in the social business, so they were particularly sensitive to the topic]

The mistake - however - is to see "marketing" as directed to business only. There are plenty of other actors (especially governments, local authorities and schools) that have actually already switched for exactly those reasons.

Most notably, the adoption of GNU/Linux by public institutions (especially schools) means to educate the future employees of the companies to the use of GNU/Linux, and this means those company will have to invest less in a switch from windows to GNU/Linux. Actually, the system is more complex than that, but it is interesting to see how much effort Microsoft and MS foundation are pouring in "pushing" windows in schools and govs...

The point of my previous post was not to deny the economical model behind FLOSS and its importance [that one is essentially self-evident (ask Google or Amazon!)].

My point was to stress that if we speak about marketing the best strategical (!= tactical) option is to point on the social/political/ethical dimension of FLOSS, because in doing so we are building a long-lasting culture, rather than an opportunistic trend.

[stchman]

http://itmanagement.earthweb.com/netsys/article.php/3672516
this guy seems pretty realistic and well-intended to me...
what do you think?
(I really want Dell to have succes in their task...Linux installed on their boxes...you know all the story...)

What I believe that a lot of Linux users want to see are "Certified Linux" laptops and desktops. Dell should continue to sell laptops and desktops with Vista but tell the customers that this machine is certified to run the following Linux distributions.

This way a person can purchase a Dell and then pop in their favorite distro and away they go.

That is what I did with my laptop. Everything on my laptop was supported by Linux except the modem and I have not used a modem in years so it was not a factor.

I looked around for laptops that had Linux pre-installed and they were more expensive than Dell, Gateway, HP, etc.

A Linux certified laptop/desktop would have Nvidia or ATI video cards, Atheros wireless, Relatek ethernet, Intel chipsets. hardware modems, etc. Devices that are known to work with Linux.

Dell, just make Linux certified laptops/desktops. We Linux folks can do the rest.

[StarsAndBars14]

Huh, I just realized...Microsoft already leverages Dell to pre-install Windows. So, since Microsoft now wants to compete with Linux by using Novell, Dell will most likely only install Novell Linux as requested, which, as I stated in my previous response, will probably end up being some kind of bastardized hybrid Win-nux thing that might end up giving Linux a bad name. (Which, for a long-term campaign to mess up the name of Linux so folks won't want to use it in favor of Windows for the next 5 years may be worth the cost for Microsoft...never know. I wouldn't agree with that kind of "unethical" practice, IE: producing a Linux OS that sucks so everyone things Linux sucks and runs back to your main OS..Windows..

???

Dell has not offered nor has it been offering Novell systems, but RHEL, which comes on their Precision xxxn line. Most systems they have placed into an "Open Source" category come with FreeDOS and you can install whatever else you want in its place. I don't see them switching from that any time soon.

A little less conjecture and a little more fact, please.

[Tundro Walker]

What I believe that a lot of Linux users want to see are "Certified Linux" laptops and desktops. Dell should continue to sell laptops and desktops with Vista but tell the customers that this machine is certified to run the following Linux distributions.

Excellent point! Some folks switching to Linux (myself included) get the growing pains when something that used to work fine in Windows doesn't under Linux. The natural tendency is to blame Linux.

"Linux doesn't support my hardware!"

No, your hardware vendor doesn't support Linux. And that's where the keystone comes into play. With the amount of free communication the net has caused, and the leap in programming technology and techniques, building a good OS isn't the stopping point anymore....it's the hardware trying to run it. Afterall, if any of us don't like Windows, we can switch to Apple or Linux. If we don't like Linux, we have the benefit of being able to alter it as we see fit (granted, there's years of programming knowledge needed, so the learning curve is beyond the average bear, but my point is...) We can directly control the software. But we can't directly control the hardware. If I don't like a CPU because it won't work with Linux, sure, I can try another vendor. But what if all CPU vendors don't support Linux? It's not like I can just go into my garage and whip one up. (Well, I guess if you're a really great engineer, you can piece together a bunch of smaller ones to make a merged chip... Sure, Intel has the Duo...but I've got the Quadra-Heli-Penta-Uber-o...it's 2000 8086 chips daisy chained together. Technically, it shouldn't work, but it does, and it provides TRUE multi-tasking...at blazing 2000 x 8mhz speeds!)

So, yeah, the hardware is a pain. I can't control that except with my dollars, and the Windows users seem to out-vote me with theirs, because there's so many more of them. I've learned to steer clear of Logitech and Creative products, which is a real shame, because I considered them to be good stuff when I used Windows. But, alas, some of the stuff I use from them doesn't work with Linux.