Michael Dell Sees Future In Linux Desktop 156
Robert McMillan writes: "Linux Magazine has just published a pretty interesting interview with Michael Dell -- not exactly widely considered to be a Linux booster. But he is keynoting at LinuxWorld in San Jose tomorrow and he does talk about why Dell is now interested in Linux. Interestingly, he also says he sees good things for Linux as a desktop OS."
Re:Building a desktop OS from scratch (Score:1)
Telakin.
Re:Building a desktop OS from scratch (Score:1)
Is there a word for simultaneously misquoting and misattributing?
Anyway, I think P.T. Barnum was one of our finest presidents for leading the slaves to the Great Egress.
Sounds like glazing to me. (Score:2)
Dell: Yes.
So, Mike, which is it? Did you spend time with some Linux dev guys or did you meet The Man himself?
Hmm.
Rami
--
Re:Building a desktop OS from scratch (Score:1)
I haven't used a mainframe for over ten years.
Sorry!
But he/she did say "Unix was designed to run on mainframes" and that sounds like he was talking about Mainframes when they were just a bunch of dumb terminals attached to a centralized unit.
Steven Rostedt
Re:Open Letter to Michael Dell (Score:1)
You could use your browsers GUI prefs dialog to disable java and javascript just prior to accessing your porn or sleazy commercial sites and then restart 'em when you move on to other pursuits...
Just a thought and YMMV...
Re:Still not easy to find on the web site (Score:1)
Re:Building a desktop OS from scratch (Score:1)
OTOH, it's cheap and easy to get hold of. There's loads of support available on the Internet and there are lots of applications available which are also cheap or free.
On the desktop, Linux has a big achilles heel in that you don't always have the latest hardware support. Hardware manufacturers put a lot of effort into writing drivers for Windows as 90% of the units they sell will end up in Windows boxes, but with Linux you are reliant on somebody in the Linux community coming up with a driver unless you are a systems programmer.
When I got my new Dell CSx laptop, it came preconfigured to install Windows when I first switched it on. Ok, the process requires four reboots and takes about an hour and a half, but at the end of that the PC and all of its peripherals were functioning perfectly.
Then I came to install RH Linux on the same laptop. The install took about an hour with two reboots (the first one being so I could restart the install in text mode because the RH install usually starts an X-server so that it can look pretty, and the X-server refused to talk to my graphics card). However, at the end of this I had no GUI. XFree86 absolutely refused to do anything with my graphics card as it was too new. I estimate I took about 16 hours or more to get it working. Is the average person going to spend that length of time getting their PC to work? I don't think so. He/she'll format c: and reinstall Windows.
However, in the corporate environment where the IT dept can control both the hardware standards and Linux config, I think it would make an extremely good desktop OS.
Re:What bothers me about Dell (Score:1)
Re:Support problems, Hope M. Dell reads this. (Score:1)
I see what you are saying about names, and I did not do it thoughtlessly.
My post would have no potential to have any sort of positive impact without this information. I honestly believe that no one
else will be able to figure out who these people are.
Said another way, I would be okay with using full names if I were saying this stuff to anyone who could possibly figure out
who they are, based on the abbreviations I used.
Frankly, I truly believe that these people are harming Dell by there incompetence (or possibly failure to follow M. Dell's
direction) and, beyond forcing me out of the job, the are harming me, and my fellows, by dragging the stock down. This
being the case, they need to be called on the carpet. Clearly they would have to be called by name.
-Peter
Re:Open Letter to Michael Dell (Score:1)
You can find lots of linux hardware here [dell.com] - preconfigured with Linux and with support
There is Linux links on the sidebar in the biz sections, but not in the consumer section.
Open Letter to Michael Dell (Score:2)
As a matter of fact, I tried to look at your web-site yesterday. But not only did it not have any obvious Linux-system links, it was configured to work with Javascript-enabled browsers only.
Not only do I regard active content such as Javascript to be a security risk, I have also had too damned many denial-of-service attacks disguised as Javascript web pages!! Too many web designers regard it as an opportunity to "one-up" frame-bombs. For this reason, I categorically refuse to run Java or Javascript.
So either clean up your act or else forget about me -- and my friends who consult me -- as a customer.
Why does Dell like Linux? (Score:1)
I like linux. But I don't think companies are going to be able to get used to it easily enough. Fortunately Dell is a hardware vendor, so they don't have to worry about open source. But still, if he says things about linux (while not offending MS), then people will like him. It's publicity, nothing more.
Don't dis M$ (Score:3)
Q. And what about MicroSoft? How will Linux affect them?
A. I hope that piece of crap OS get's plowed under like last years chaff so that we don't have to support Gate's business model anymore. I'm so sick of supporting his closed source flotsam I could puke. Linux will let us cut our support staff in half, and with the source code we'll be able to push the envelope of what the PC can do. In a few years, you will see us shipping Dell computer that truly inovative and not just Wintel tag-alongs.
Re:What bothers me about Dell (Score:1)
-={(Astynax)}=-
How can Wine "overtake" Windows? (Score:2)
How can Wine's Direct3D emulation (or Wine itself, for that matter) "overtake" Windows? It is trying to BE Windows. Can Wine really be a better Windows than Windows?
Re:Building a desktop OS from scratch (Score:1)
Linux can please all of the people all of the time, simply because there is more than one distribution of it. If you want servers, go with Debian, if you want desktop, go with Mandrake. I wouldn't use Mandrake to host a website getting 300,000 hits a day, but I will use it to play my mp3s and play games.
Converting Linux into a desktop OS is the most feasable alternative to Windows right now, if that's their aim. And the fact that there are different distributions of Linux insures that if it dies horribly, that it won't take everyone down with it, and if it succeeds, that Linux will still be able to please those who like to hack the command line and those who like to point and click.
- Wedg
Re:Building a desktop OS from scratch [OT?] (Score:1)
I have a dream where you walk up to a computer (at that time rather a terminal), you touch it - and the usage of it is as evident to you as the usage of a hammer.
Mark Weiser at Xerox PARC had the same dream of ubiquitous computing. Many small computers embedded around us that function invisibly without the user having to worry about OS, software, hardware, etc.
PDAs are close, but they aren't transparent enough to us because we still have to fool around with the clumsy interface.
Tivo is closer, is acts like an appliance should-- hiding the underlying operations and recording my shows. I don't have to worry about filemanagement or any crap like that in my VCR, it just works.
In the morning, I just want to turn the knob for darker toast, I don't want to tweak the toaster kernel and recompile...(meanwhile on my PC, I enjoy the tweaking)
Back to the point: I agree that OSes should shift to make our computers more like tools, but I don't think Linux and MacOS are particularly wrong for it.
____________________________
wha (Score:1)
any OS that is that hard to use is not meant for the desktop.
i think the biggest thing holding it back is the inconsistent interfaces... linux may be OK for geeks with nothing better to do but like, script Perl or whatever, but for people who actually have to USE them to GET THINGS DONE NOW, linux is not an option.
mmkay? excuse my blasphemy.
Re:Building a desktop OS from scratch (Score:2)
I'd suggest that instead of removing command line and multi-user functionality from the system, what you really want to do is to hide it. NextStep did at least part of this fairly well, by hiding the "normal" Unix directory hierarchy, because no user ever really needed to see it.
Similarly, I'd suggest hiding the existence of different user permissions, by running as little stuff as possible as root, and never allowing root logins, but making all manipulations of the (hidden) system files dependent on a concealed "sudo". This prevents users from accidentally fubaring the system (I assume you don't always login as root, for exactly this reason), while meaning they don't have to know anything about users or permissions beyond entering their name when they start the system.
Re:Building a desktop OS from scratch (Score:2)
I'd have to disagree with you there. Yes, NT fails miserably on all of those, which makes Linux look good. However, that doesn't mean Linux is lacking in those areas. In fact, it does pretty well for all of them. I'm not claiming it's the best in any area, but it's far from the worst. It doesn't scale to the high end as well as Solaris, DG/UX or Irix, for example. Nor does it have the reliability of Tandem or the security of the various MAC-enabled OSes. However, for probably 80% of businesses, it's a suitable server OS for their needs. It has sufficient scalability, reliability and security to get the job done.
Not to sound hypocritial, but..... (Score:5)
"Interestingly, he also says he sees good things for Linux as a desktop OS."
Yet yesterday, 90% of slashdot readers berate AOL for providing their suite of access products for Linux
What's it to be ?
Interesting. (Score:4)
The other thing is -- by supporting Linux, they rather seem to be competing with SUN & al. then with Microsoft. A lot what he says remains unspoken (e.g. he says "Look at SUN and Microsoft", and then doesn't even mention the latter). Linux warriors may be more obsessed with Microsoft, but one conclusion I draw from that interview is that rather the big commercial Unix corporations can be the real competitor / enemy (however you state that) of Linux in the corporate market.
Best regards,
January
Re:Games = desktop success (Score:1)
Re:Games = desktop success (Score:1)
"Direct3D emulation under Wine is a little behind"
hmmn, did you fingers slip on the keyboard ?
Shouldn't that be Wine is a little behind ?
Re:Building a desktop OS from scratch (Score:1)
Re:Of course they see linux desktops in the future (Score:1)
OK, let me start by saying I love using linux, I think it can be used on a desktop if a few things happen here and there, regardless of its initial intent. I'm even using it NOW.
FYI, linux was not designed to be a server, it was designed to be a hobby for a grad student. I'm not saying that yout statement is wrong, I'm just saying Linux is already being used for more than it was intended, so why not expand it to be used the way people want to use it (ie desktop)?
As far as taking server market away from Sun. If money was not an issue, I'd MUCH rather have a Solaris server. If money was an issue, I doubt I'd be able to afford one of Dell's expensive-as-hell machines anyway. If I was tight on money, I would be able to put a machine together for MUCH less and make my boss just as happy or happier.
To quote the Jargon File... (Score:3)
... any noun can be verbed.
A hint for Dell (Score:2)
Re:Games = desktop success (Score:1)
Uhm, nope. If Joe Sixpack wants his kids to play games, he'll get them a PlayStation or DreamCast just like all of their friends have.
Nope again. Games existed for DOS/Windows long before DirectX and related goodies in huge numbers. Games are written for Windows because of the huge installed user base. Developers put up with DOS's terrible memory model (I was about to say memory management but reconsidered) and built drivers for every sound and graphics card from scratch for every game or line of games. They put up with this because people used DOS and were going to buy games for DOS, and didn't care much about how easy it was to develop for it.
The availability of "productivity" apps, like an office suite, is a much more important factor in swaying users to Linux. Conquering the home market is hard, and not the first priority. Conquering the small- and mid-size workstation market in the office is a much more likely target.
There's no real commitment here (Score:1)
Re:Building a desktop OS from scratch (Score:1)
Re:Building a desktop OS from scratch [OT?] (Score:1)
Actually I think it is, completelly.
Ubiquitous computing needs a paradigm shift. No filesystems, no procesors, no configuration files. Forget it. It just works. Somehow. It is simply stored. Somewhere. It runs. Who cares whether it's stored at your drive or on a fileserver in Japan. You only expect the data to be available to you (and fast) regardless of you being in New York or in Delhi. The same aplies to CPU power (which can be quite revolutionalized by things like ProcessTree), configurations (wisely estimated from your previous options in other programs you used elsewhere) and many other things. And I am quite sure when ubiquitous computing will start being take seriously this will not even scratch the surface of desirable behavior.
To the points that others have raised -- about computers being orders of magnitude more complicated then hammers and toasters -- well, no doubt they are. Many things in life are. But they can be made so easy you do not know about it. Music CD player is a really advanced peace of laser-optics physics, quite a lot of electronics (audio filters etc.) -- and you do not know a bit of it when you use it. The point with the car was actually quite good -- yes, you do have to learn a lot before you drive a car. But you are actually required to, because by driving a car in a wrong way, you are probably going to cause much more damage (including damage on lives) then with a mere rm -R / on your system. With that, you are only going to hurt yourself. If cars would be the same, not many people would take care to learn driving actually. You just sit, turn the wheel and press some pedals -- and hopefully it will move.
Re:Interesting. (Score:1)
Free markets are a good thing; and market driven businesses like Dell are ideal in a free market. Technology-driven as opposed to market-driven businesses tend to fail anyway, for the simple reason they aren't always concerned about offering what people want to fork their money over for.
Linux on the desktop and Dell (Score:2)
Will Dell just give you the option and not help you if the OS and the hardware don't play nice or will they back up the kind words with corporate action?
Re:ummm... get WHAT done now? (Score:1)
Re:Not to sound hypocritial, but..... (Score:1)
Yet yesterday, 90% of foreign car owners berated German-made radios in their cars.
What's it to be ?
The point: these are completely orthogonal
It's worse than that (Score:1)
-jpowers
Not to put a downer on things (Score:5)
Out of all the available Operating Systems out there, how many are based on just two standards? There is Win32 (Windows 9x/ME, WinNT), and POSIX (All *nix variants, most BSD's, BeOS, QNX etc). Only two standards? Where is the choice? Where is the inovation?
It seems to me that all anyone is interested in is twisting and squeezing Linux into ever more bizarre and improbable situations (Linux on palm tops, Linux for embedded devices, Linux for games consoles etc.) Is the market stagnating, where no one dares break away from the pack and try something new for a change?
Maybe it's envy on my part, but why can't we have an Open Source project that isn't based around some form of Linux or a POSIX kernel? Is there any room for innovation in OSS these days?
Just had to get that off my chest, sorry.
Re:Not to sound hypocritial, but..... (Score:1)
"Interestingly, he also says he sees good things for American-made motors in foreign cars."
Yet yesterday, 90% of American-made motor owners berated German-made fan radiators.
*What's it to be?*
Not everyone is a fan on Dell it would seem... (Score:3)
Re:Building a desktop OS from scratch (Score:1)
Um....I'm not sure I follow. First, Linux is a kernel, nothing more. Repeat after me: Linux is a kernel, nothing more.
Second - Builing the GUI into a kernel has some severe disadvantages. Just ask the users of Windows about any stability issues they might have. Many people view this as a bad thing. Perhaps a more modern X server that is designed for the needs of the modern desktop and modern hardware, yes that might be a good thing.
Third - No CLI.. uh... why? I'm typing this response on Win2K and guess what... I'm just a few clicks away from a CLI (they call it a Command prompt now, no more Dos prompt)and it's very usefull. Do I have to use it to get around Win2K..no. Do you have to use a CLI to get around Linux...no. You can configure X to start up when you boot and you can do whatever you want to do without ever looking at a CLI if you so choose. But again, why would you want to eliminate the choice. That makes zero sense.
Fourth - Your premise is that by following your bizzare specs for this new desktop os you will have no feature bloat. Uh...why? Feature bloat is everywhere I look, Gnome, KDE, windows, etc... I'm afraid that it's our nature and not really a function of a specific desktop environment.
Fifth - JFS already ships with Mandrake 7.1. You are welcome to use it if you wish. Also, not to be anal here but a JFS is typically needed more for the server arena than the desktop anyways.
sixth - Many of the hardware support issues you speak of when you refer to PnP and "advanced multimedia" support can be directly related to the total lack of vendor support for device drivers. I'm not sure how your new desktop OS plans on dealing with these issues or if they will simply dissapear like magic.
Instead of tossing the baby with the bathwater why not continue our efforts with Gnome and KDE as they seem to be making great progress towards becoming viable desktop solutions running ontop of *nix.
Re:Building a desktop OS from scratch (Score:1)
I don't believe it's that hard to have some configuration tools that are easy enough for the beginner. The other necessity is a good unified environment, things look nice on the gui front (KDE and Gnome are quite good, but perhaps it would be good if one won over the other and we had ONE desktop for newbies), but there are problems in the inner works...
We have some sort of distro hell here, because different distributions work in different ways, put their config files and things in different places, and that is a big tech-support problem. I used to work on this and, believe me, it's good to tell someone to do start->run->winipcfg to learn about their IP address.
Linux needs some standards, at least for config files and directories. One simple way to set up dial-up, for starters!
I don't think we need more OS's, though I would welcome them; laying my hopes on the new Amiga, even; but it's quite difficult to catch up on the established-I find it surprising that Linux has come to this level of popularity.
[OT] Let's start searching! (Score:1)
Dell is our friend :) (Score:1)
not much of a true believer (Score:1)
not much of a true believer is he?
Re:Building a desktop OS from scratch (Score:1)
I would agree that rather than trying to make a handheld run X effort would be better directed toward app development, but that is why it is called the open source community, not open source inc..
Maybe take a look at Plan9 from Bell. BeOS is also good but you don't hear much about it here because it is closed. At some point an OS will emerge that has *really* innovative features rather than repackaging old concepts. I doubt it will be instantaneous process, likely a series of improvements along the lines of BeOS or Plan9. (or like X a few years ago)
Re:Building a desktop OS from scratch [OT?] (Score:2)
OSes are still made by a bunch of programmers for a bunch of programmers. And the computers are now SO pervasive and the economical loss resulting from bad usability of these tools is so big that I believe (hope) this cannot last long.
Tool was the important rediscovered buzzword in the last paragraph. Because tool is all that computer is. It's not a piece of art, it's not a personal friend -- it is a mere tool. Like a hammer or a pencil or anything. Only a bit more versatile.
The main OS shift is probably happening today with PDAs and mobile devices. They are the first widely used single-purpose computer technology base TOOL. And more functionality will migrate from desktop PCs to tools. Bluetooth will help along this line.
I have a dream where you walk up to a computer (at that time rather a terminal), you touch it - and the usage of it is as evident to you as the usage of a hammer. (No, it's not going around and bashing things with it.
Michael Linuxes for business... (Score:3)
LM: Did you actually take a look at the technology?
Dell: Yeah. I have a little lab next to my office here, and I got a desktop PC and installed Red Hat Linux on it. I played around with it a little bit.
...and...
LM: OK, how much time do you personally spend thinking about or dealing with Linux these days?
Dell: I don't really have a number that comes to mind for you.
I bet he didn't even partition the hard drive himself. Granted Dell is a hardware supplier, I don't need its CEO to be a Linux efficianado, but some expressions of capability at the top sure would convincing.
His comments sound like many others these days: "We're going to support Linux because everybody else is and we don't want to be left out. But trust us, we know what we're doing."
Hardly "a new Linux evangelist" in my book. Our community needs companies that develop business units whose sole function is to support us (Linux), not just have us tacked on to rest of the OS support department. I think we'll see that the most successful companies five years from now will be those that got in to Linux with both feet today, provided dedicated service, knowledgable expertice, and serious committment from individuals that are compassionate of the cause. If anybody can spot an imposter, its going to be us.
Linux hasn't been easy until now, and we sure aren't doing this on a lark. How about showing some interest before you ask for our money? Sorry Michael, I'm not convinced.
Re:Michael Linuxes for business... (Score:2)
I'm sorry, but I can't see why this is a problem - Dell owns a *very* large computer company; that he retained control when his company GOT to be large is commendable, that he still has a little "play area" right next to his office is almost unheard of.
However, the end result is DELL computers will ship with Redhat and full support. This *is* a business decision, not a geekish one, so how else can you expect him to treat it?
I bet he didn't even partition the hard drive himself. Granted Dell is a hardware supplier, I don't need its CEO to be a Linux efficianado, but some expressions of capability at the top sure would convincing.
Well, he claims to have installed Redhat himself (I suppose he could have just overwritten the existing partitions set up by a more competent geek, but odds are good he started with a "clean" unpartitioned disk and let the Redhat bootable CD do the partitioning) and redhat usually makes some reasonable default decisions. However, again, I don't see how this is relevant - He got a Redhat up and running himself, played with it for a bit, and was willing to give an interview praising Linux and advertising that his company feels it is ready to be shipped pre-installed.
Our community needs companies that develop business units whose sole function is to support us (Linux), not just have us tacked on to rest of the OS support department.
Why? few if any companies have a separate support department for NT/Win9x/OS2/Novell so why should they have a separate one for Linux? one of the *big* points that is always made about Linux is that it doesn't need as much support, and that *that* support is often better found from the community anyhow. However, even if that *was* true, you have obviously failed to read that part of the article carefully, as you missed:
which is pretty much waving the "get your dedicated linux-os support like this" flag at you...
Linux hasn't been easy until now, and we sure aren't doing this on a lark. How about showing some interest before you ask for our money? Sorry Michael, I'm not convinced.
I am sure MIchael will have troubled sleep knowing you don't want his products.
--
Re:Not to put a downer on things (Score:2)
Want innovation? Have a look at the Tunes [tunes.org] project.
Re:Building a desktop OS from scratch (Score:2)
The purpose of the OS is to control access to parts of the computer by other parts of the computer. When you sit down at the PC, you can consider yourself just another input device. Why shouldn't the OS control input? This is done through multiple logins.
Now a personal computer would be 'personal'. It would only ever have one operator. But shouldn't the OS still protect your personal computer from unauthorized access? Unless you're planning to have the thing embedded in your brain I can only see requiring a login as a benefit.
Get rid of the CLI? The most productive interface available once the learning curve has been overcome? Why?
I'm glad that you have a perfect power supply where you live so that a journalling file system has no advantage, but the unfortunate fact is that the rest of us live in the real world. The journalling FS is there to protect data. I can see how it would be unnecessary to protect your saved-games, but some of us use our destops for real work.
The GUI is an incredibly heavy piece of software that is near impossible to prove correct with todays techonology. Wrapping it into the OS is stupid. What you propose is to take the most important part of the computer and wrap it with instability. Just because I'm the only person using the computer doesn't mean that I don't have jobs running in the background.
So if we design a new OS, we want have problems with licensing problems or feature bloat? Some people decided to re-write Netscape, a mere application. What do people complain about?...licenscing problems and feature bloat (whether they actually exist or not).
After replying to your post, I get the feeling that you are a troll. "Heh everybody, dump Linux and use something that looks exactly like Windows instead!" Do a little research to find out the advantages of the 'UNIX way' before posting that we should dump it.
Support problems, Hope M. Dell reads this. (Score:2)
other person in the department competent to do level 2 support for Linux quit.
Dell server support has plenty of Linux "boot camp" graduates, or people who suddenly realized it is "cool,"
but all of the real experience was driven out of the department. I, really, by Michael J. and I suspect my
friend really had to leave because of Gene B.
I hate to post this, because I am a stock holder, and I am losing my ass right now, but Dell is driving good
people away because they don't "fit the mold." I want to see Dell succeed (because when Dell does well I
make money) but I couldn't, in clear conscience, recommend a Dell server with Linux factory installed
because there is no one to support it.
It seems that today Dell as all "alliances" and no substance.
-Peter
Dell's Linux support can be testy at times (Score:1)
A few months ago we got a Dell server with Red Hat preinstalled. To our surprise, the "Linux" compatible system was "Linux ready". But it had a binary-only RAID controller driver, which caused much grief amongst us when the recent kernel security hole (found by the sendmail project people) was found. We had no good way to upgrade the Linux kernel to close the bug without losing access to our drives.
Note however, that this may have been Adaptec's fault; they supposively designed the RAID system, and may have been trying to keep its features private. One could also have blamed us for not researching the "Linux compatibility" further. But we were scared for a while since we heard rumors that they were having problems getting the driver to compile with the new kernel.
Since then, they have released an update (over a month after the hole was first reported), and it includes an open-source version of the binary driver, but we have to wonder what was going on in their heads when they tried such a move. The driver (aacraid) is in Red Hat's Linux's current source code as well, but I don't see it in the mainstream (or for Linux 2.4.0-test#) yet. Drivers in Linux tend to be tied to the kernel; if you don't have the source to upgrade them, you lose the compatibility. Dell only learned that their common usage of custom hardware and drivers wasn't going to work with Linux when a whole bunch of us starting screaming on their message boards.
Re:Dell is our friend :) (Score:2)
Re:Building a desktop OS from scratch (Score:1)
Create lightweight desktop applications not with configurability, but simplicity in mind. Avoid redundant functionality (i.e. button bars doubling menu entries). Most of these apps are already there, just port them to a common toolkit (fltk, for example).
I'm not sure why you threw in "avoid redundant functionality," but I have to point out; duplicating functionality where convenient is a common (and wise) practice in interface design. The point is to have a consistent, easy to find location for a function (i.e. in a menu), but also have it convenient when needed.
Reduce inter-application interfaces to classic Unix pipes, sockets and libraries. Avoid bloat and slowdown through Corba and similar interfaces.
Which causes more bloat and slowdown, a component-based system where common components can be shared across applications when needed, or a system where every program needs to build its own functionality from libraries? So, what, we embody an entire HTML-editing component and interface in a library? We increase development (and package management) headache exponentially for... what?
As for pipes and sockets, I can't wait to hear how your GUI OS makes these easier to use than drag-and-drop embedded components. Or how you explain pipes and sockets to your grandmother.
Re:Building a desktop OS from scratch (Score:5)
- Throw multiuser and files access permissions out of the system. have the user automatically login and work as 'root' just as in windows 95/98/me and in macos. sure, this creates a lot of security issues, but they could be tolerated on desktop machines with dialup-only internet access. by far more simple for unsophisticated users.
- Simplify the file system structure. If multiuser functionality has been removed, it is only longer necessary to have
/home, /sbin, /usr/sbin and to have both global and user-specific configuration files. Join /bin, /sbin, /usr/bin, /usr/sbin, for example. Rename all system directories so that they are easily understandable in natural language: /usr/bin to /Programs, /etc to /Programs/Settings /lib to /Programs/Data. Create /Fonts, /Sounds, /Pictures, /Documents, and so on.
- Throw out X11 - since no desktop user needs its network functionality - and replace it with a framebuffer-based GUI.
- Create lightweight desktop applications not with configurability, but simplicity in mind. Avoid redundant functionality (i.e. button bars doubling menu entries). Most of these apps are already there, just port them to a common toolkit (fltk, for example).
- Reduce inter-application interfaces to classic Unix pipes, sockets and libraries. Avoid bloat and slowdown through Corba and similar interfaces.
- Adapt and simplify LinuxConf to act as the system configurator
- Standardize on one scripting language in your "distribution" (for example, Python). Avoid that several bloated scripting languages have to reside on the system just because system utilies (package managers etc.) need them.
I agree that the resulting system will have litte in common of what we know and appreciate in GNU/Linux. But it would be the perfect system for people who don't want to replace the complexity and impenetrability of Windows with yet another type of complexity and impenetrability. Hammering nice graphical interfaces on top of that complexity won't help.Company man in "Make Money" shock! (Score:1)
(Paraphrasing) 'Our customers want to use linux, so we're getting around to making it easier for them.'
His bottom line: to sell machines. Good for Michael Dell, but is this actually news? It's not as if he's actually annouched any big concrete push to support linux or anything...
best wishes,
Mike.
Re:Building a desktop OS from scratch (Score:2)
I'm visualizing a system which is quite independent of anything else - new apps and games would have to be written specifically for it, and installation of apps would be handled from within the GUI via an installation wizard. These new programs would avoid all the headaches associated with installing software on Linux - they should be installed in a directory under the GUI system's directory, and depend on a new, stable set of libraries, to avoid problems with dependencies. Also, all programs would follow a pre-defined look-and-feel, support the same cut-and-paste features, etc. Make it cohesive from the start.
This GUI system would be written from the ground up to be an all-in-one desktop solution. X-Windows has a lot of cool features, like network transparancy, but it is too complex for the average user, and developement for it is also complex.
This would still be a monumental project, as it ignores all existing software, but it would be a lot more feasable than writing a new desktop OS from scratch.
Comments?
Re:Building a desktop OS from scratch (Score:3)
I DO use Linux as my desktop OS (both at work and at home) and I am very happy with it while Win9x (the supposed REAL desktop OS) keeps constantly annoying me.
My mother (well, she is not exactly a hardcore computer geek) uses Linux as her desktop OS. She is quite happy with it.
And anyway, I just see no reason to NOT to use Linux as a desktop OS. Could you mention just ONE feature that shows that it should not be used for that?
(Remember, in the old DOS days people whined about having to shut down Linux as opposed to the 'just switch it off' method of DOS - and see, what happened in Win9x)
Re:Dell Has Been Raymondized (Score:1)
Dell: The question really is does Linux create new users or does it take users away from Sun or Microsoft. I'm not sure I'd know the answer. To be perfectly honest, I'm not sure I really care, as long as they use Dell.
But it's nice to hear someone in his position recognize the " ... powerful legitimate shift."
Re:Building a desktop OS from scratch [OT?] (Score:1)
Rather than focusing on the tired examples of 'hammers' and 'toasters' - both which take arguably no to minimal intelligence and skill to use, treat computers like they should be treated - like cars.
You need to learn how to use one, learn the idioms behind the design, and have some experience using them before anything about them become self-evident.
--
Paradigm Shift (Score:1)
So what are our other choices? OS/2, BeOS, Freedows, MacOS, DR-DOS?
OS/2 great ideas, IBM did a good job on designing the OS and picking up where Microsoft left off. The problem ends up being third party support, or lack thereof. Once even Wordperfect bailed from the OS/2 platform, I knew things would not go well for the OS.
BeOS, the AmigaDOS of the 1990's. It has its niche market, but still has a long road to walk down before it gets more marketshare. Don't get me wrong, I like BeOS, its a fresh new technology and it is powerfull, just not popular enough to get the third party support behind it needed to get adopted by other PC companies.
Freedows, not ready for prime-time. I am not sure when they will even get an Alpha test out. Is that cache kernel done yet?
MacOS, this is going away once OSX comes out. The problem here is that it only runs on the Macintosh Platform, the most recent PowerPC Macs that is. The old 68K Macs can't run it, and there are no plans to port it to the WINTEL platform like BeOS was ported. All I see OSX as is just yet another BSD Unix hack. Might as well use BSD on PC systems then, you may not have that Aqua interface, but at least you can have multiple platforms BSD can run on and have the scalability that Mac hardware lacks.
DR-DOS, the OS that refuses to die. How many hands did DR-DOS pass over? DRI, Novell, Caldera. I don't really see this OS doing much, unless a majority of the market rejects Windows and its GUI and gets back to basics to run DOS programs.
BSD Unix (FreeBSD, OpenBSD, NetBSD) would be a good alternative to Linux and Windows. I don't hear as much Hoo-Haa about BSD as I do Linux from the PC Makers. I guess maybe that GPL license makes the difference, or is it that Linux gets most of the good press from the media?
Re:Building a desktop OS from scratch (Score:1)
Bad idea. Even on a simple dialup PC, your chances of getting hacked while online are non-zero. Besides, what if more than one family member uses this box? How do you protect your own files from accidental, never mind malicious, damage?
Simplify the file system structure.
Why? Why not just make the interface smart enough that the user doesn't need to know about the filesystem structure? This is slowly but surely happening anyway.
Throw out X11 - since no desktop user needs its network functionality
I'm a desktop user, and I need its functionality.
Create lightweight desktop applications not with configurability, but simplicity in mind. Avoid redundant functionality (i.e. button bars doubling menu entries).
Configurability is a good thing, as long as it doesn't get in the way of simplicity. I like highly configurable applications, as long as the defaults are sane. Redundant functionality is a good thing: different users have different preferences about how they want to do things.
Reduce inter-application interfaces to classic Unix pipes, sockets and libraries. Avoid bloat and slowdown through Corba and similar interfaces.
Thus crippling any kind of useful component object model that Unix desktops are just truly beginning to take advantage of.
Adapt and simplify LinuxConf to act as the system configurator
Having a useful system configuration manager is a desirable goal. I pray that LinuxConf is never adopted as a standard though - throwing it out it what's really needed. Maybe that's me, but I truly loathe LinuxConf.
Standardize on one scripting language in your "distribution" (for example, Python).
Argh. Please no. Allow developers and users their choice of what scripting language they want to use. Sure, Python's a good language, but why Python? Why not Perl or Tcl or Scheme?
To be honest (and I don't mean to flame you), it sounds like you'd be happier using something like BeOS or MacOS. Why go to the trouble of using a complex, powerful Unix-like kernel if you're immediately going to rip out half of its functionality?
Re:Building a desktop OS from scratch (Score:1)
Throw multiuser and files access permissions out of the system. have the user automatically login and work as 'root' just as in windows 95/98/me and in macos. sure, this creates a lot of security issues, but they could be tolerated on desktop machines with dialup-only internet access. by far more simple for unsophisticated users.
>>
You know, this is one reason I use Win98SE almost exclusively
(well, get to gaming, rather
Eric ze Kidder
Re:Building a desktop OS from scratch (Score:2)
Actually no. Linux's sole purpose is to give Linus Torvalds a Unix that he can run on his PC. That's why it was born, and any other uses are purely coincidental. As it happens, Linux is a pretty good server OS. It is also (and here's the controversial bit) the best desktop OS in the world. No, I'm not on drugs -- I seriously believe that it is the best desktop OS in the world for me. It does pretty much everything I want, and it does it better than all the alternatives I've tried. I'm not foolish enough to claim that it's for everyone yet, but given time it will get there. The features that make it a good server OS don't preclude desktop use. Or are you going to try and claim that Win9x is a better desktop OS than NT4 or Win2K?
Re:Building a desktop OS from scratch (Score:2)
You do have valid points, but I must say that the problem may not be with the OS but what we are doing with it. We say that Unix/Linux is not for your grandmother, and I would agree. But *nux is very good with networks. Now the problem is, we are trying to get Grandma onto a network. The Internet.
Now this causes all sorts of problems. I saw one poster a few days talking about how their wife complained about having to log in. "I own this machine, and I'm the only one on it". But that my not be the case if you have an Internet access. You see when you take single user methodologies and put them with network ones, you get things like virus and privacy compromises.
So, we need to educate the average user and maybe tweak the OS a little. But I really think that if you are going to have a connection to the internet, you should have basic knowledge about how to use a computer. The normal analogy is to compare computers with cars. You may not know how a car works, but you definitely know what to do and not do with it before you drive. That's why we have licenses, (although I would say there are those that don't know how to drive). A computer is no different (except for being less dangerous). You should have a basic knowledge before surfing the net. And I'm tired of hearing how Joe Schmoe doesn't want to know anything before using a computer. If all you want to do is play games, that's fine, but if you accept email and go out onto the net, then either learn or accept the consequences.
Steven Rostedt
Innovate it however you want (Score:3)
Want to add obscure nifty features like ACLs and Roles? You can do that, kind of like the LIDS guys and the Linux ACL project.
Want to rip underlying kernel out but keep the rest of the look and feel? Kind if like Debian/HURD maybe?
Want to keep the underlying kernel and get rid of the look and feel? Like all the embedded Linux projects like the Tivo do? Go for it!
Want to do something else? You're free to.
This isn't so much a Linux phenominon as an Open Source one. People have said inventions don't happen until we're ready for them. The world wasn't ready for Open Source until everyone got wired via the Internet. Then it just exploded and companies started to realize that with the playing field level everyone can profit.
Something radically better than Linux may come along at some point (Though I rather doubt it'll come out of Redmond) and people will start switching at that point. Innovation will still happen, but now it'll be a lot more people controlling the direction we go.
Heh! (Score:2)
History of unix repeated? (Score:5)
And the companies thus arose, and they fought each other, with various wondrous types of software, none of which worked with the other. And everyone saw that there was competition, and it was good.
Thus began the ten years of drought, when application developers moaned, "Wherefore am I to write my software? None shall buy it, for each is on his own little island, each separate from the rest." And Gates said to them, "Come to me, ye fools! And ye shall be happy, for all my denizens live under one roof. It leaketh sometimes, but ye shall earn gold selling yer stuff." And the developers all flocked to Gates, and the unix companies continued to fling dung at each other, and it was all as it had always been.
w/m
Re:Not to sound hypocritial, but..... (Score:2)
There is no such thing as 90% of slashdot readers berating anything - if they did, you'd have to sift through 200,000 or so posts per article... 90% of the vocal comments on that article may have been against the AOL thing, but I'm not even sure that's true. Granted, I browse with highest scores first unless i'm moderating, but most of the posts I saw were saying that this is a *good* thing - or at least not particularly the product of evil.
Even if 90% of the comments on that article were vehamently against it, that's still only a small percentage of
So, for the sake of not being entirely offtopic - Personally I think having Linux as a reasonable desktop competitor to M$ would be great, sure, let the idiots use Linux - people will write applications for them, and we'll write applications for us. However, I don't want to see a linux machine on every desk any more than I want to see an iMac or a Windows machine on every desk... competition is the key to survival... I think it's great the Michael Dell sees a viable market, I think it's great the AOL is supporting linux, these businesses wouldn't do anything they didn't think would succeed... So maybe Linux will put a major dent in M$'s market share, and everyone will get better applications, and better prices. Yay free market economy!
-nosilA
What do you want from a desktop OS? (Score:2)
My anecdotal evidence is that my room mate's Windows machine (Which I have as little to do with as possible) seems to become increasingly unstable as time progresses. My Linux machines seem much more resistant to this, assuming you can get the software and install it in the first place. If I were to religiously install everything from RPM, I think it would be a lot more stable. The problem with Linux of course is that my room mate can't get all those games she wants and I'd end up having to set up her 3D card and networking and stuff. And not every average person wanting to run Linux can find a guru to room with.
Re:Not everyone is a fan on Dell it would seem... (Score:2)
1) This kind of business-city dealings have been going on for years, and written about in business 'zines, discussing whether it's a good idea or not. If the city thought it would be a raw deal, they could have told Dell "No" to its demands.
2) "...preferring to bring in desperate workers from non-computer fields who cannot find any other kind of work. " Those bast**ds! How dare they hire people who would otherwise be unemployed! And then having the audacity to pay them >$20k/yr in a state with a low cost of living! Really, if these jobs are truly the "third-world assembly work" the author claims, does he really believe that "sharp, well-educated people who have computer experience" would want them? No, they go for the high-tech, high-pay first-world information-age jobs. Duh.
3) "Dell is in the assembly business, not the high-tech business...Dell pretends he is running a high-tech company...But Wall Street knows better...Dell's stock has dropped from 55 to 35. This is what happens when people figure out that [Dell is just a manufacturing plant]. " This is just wrong. Investors have not suddenly realized that Dell is *gasp* a manufacturer of PCs. Hello! This company has been doing the same thing for about 20 years now. Their business model and implementation is not just now being understood. The reason the stock price has dropped is because the commodization of PCs (caused by the low RAM, HD, CPU prices) has created a significant sub-$1000 PC market, which has much lower margins than the mid- and high-end systems, which historically provided the real corporate profits. Dell has been trying to figure out, along with *every other* PC seller, how to make money from low-cost, low-profit systems. Everyone has been dinged due to profit problems in the transition from the past several months. Furthermore, there was a sympathy effect with the drop of MS stock (from the DOJ action) than impacted everyone for a while. Furthermore, investors might also think Dell is making bad decisions with this TN plant, or with the Linux moves. But it's not because they suddenly figured out that Dell builds PCs.
3b) This is a facetious argument regardless, because manufacturing is a non-trivial operation. To create a manufacturing plant with a huge capacity, very low error rate, in which every system can be customized, and the components are fragile, is a challenge. You need smart people who know know about manufacturing process, analysis, finances, and technology. I doubt Dell has a bunch of monkeys grabbing parts from bins and throwing them into cases. I would wager that they have a rather sophisticated assembly line, developed by smart, well-paid people.
4) "It's pretty obvious that Dell's slickmeisters have taken these country boys real good." Yep, that smooth-talking, well-heeled Texas boy sure did pull a fast one on those inbred, knuckle-dragging, TN red-necks. Good argument; when things go wrong, accuse the other guy of being smarter than you. Man, I hate smart people! Always using their intelligence to their advantage!
In the final analysis, the author has two real gripes. The first is that rich Michael Dell got a sweet deal from TN, and hasn't yet fulfilled his promises. Well, if the TN people are so stupid as to give mucho benefits to a company with no guaranteed benefit to themselves, I have trouble feeling sorry for them. His other gripe is that the introduction of a large business has negatively impacted the surrounding areas. That's a real issue that cities and states have been and still are grappling with for a while. But it's not Dell's fault. It's Nashville's fault. They invited them in, and now they have to deal with the consequences.
Re:Still not easy to find on the web site (Score:2)
Let's compare:
Windows:
1 man-day to install.
2 man-weeks to actually get everything working
1 man-hour to make a disk image
Linux:
1 man-day to install
2 man-weeks to actually get everything working
1 man-hour to make a disk image
Where is the extra time for Linux?
Note: Dell doesn't run an install from CD on each machine seperately. They create a master image and copy that to thousands of disk which each go into one of thousands of identical machines. If they didn't do it that way, Linux would win hands down, because you would do the install and then run "Dell's Install Script", while the Windows machine require fiddling with the GUI seperately.
Re:Don't dis M$...BUT (Score:2)
When you put together some of the things Dell has been saying lately, it's pretty clear he has problems with Microsoft, even if he cannot discuss them openly.
Dell (from this interview, with emphasis added): "I think it definitely has the potential for a lot of change -- and disruptive change. Not so much on our business models, but on other business models. The whole open source movement has the potential to really change the way value is created and distributed in the software industry -- the speed at which applications and tools are developed and deployed. And Dell's the perfect hardware platform to do that on."
Dell (from a Charlie Rose interview at a conference in Paris): "If you had a business that was based on tricking your customer -- which, in fact, a lot of businesses were fundamentally, assumed that the customer didn't have much information and the customer was, in fact, uninformed -- well, that's going away."
When you put these together, it seems to me that Dell is inching away from Microsoft because he sees a big fall in their future. And why not? Look at the MS business model from Dell's point of view:
He can remember the days when PCs sold for $3,000 and the OS cost $15. Today he sees computers selling for $500 to $800 with an OS that costs $85 to $250. And the vendor selling the cheap box has to provide support for the OS, even if it turns out to be a piece of crap.
Somebody's getting squeezed here, and Dell knows who it is (even though he probably has a really sweet deal with Microsoft). He knows the main reason he has to keep selling MS OSes is because of the ignorance of its users. But that doesn't mean he has to like it. And he sees the Internet as working against software vendors who rely on the ignorance of their users.
Sure, he dare not speak its name. But Microsoft clearly fits his description of software companies whose business models will not survive in the Internet economy. We should not be too quick to doubt his sincerity since his conversion (though late) seems to be based on solid values and genuine interests (that's interests as in business-type interests, not curiosity-type interests.
Re:Building a desktop OS from scratch (Score:3)
Actually, you don't have to remake Linux to be a desktop OS. All you have to do is put a "desktop layer" on top of Linux. That has the following advantages over remaking or starting from scratch -
--
Re:Building a desktop OS from scratch (Score:2)
Re:No M$ Tax ? (Score:2)
And what makes Linux exciting to most if us is the DIY stuff involved. If MD really want to listen to the first generation of Linux users he should take this into account and then deliver OS-less laptops (servers, etc.) instead of ones with specific OS installed.
How much money is MD expecting to charge his customers for a Linux enabled laptop ?
Let's evaluate
- Internal R&D rentability, including eventual driver development and staff training
- Technological survey and distribution optimization
- possible prices raises on the Windows-powered side as Bill might raise his prices because of such a lack of fidelity. As MD didn't especially speak about a divorce, he might not want to lose its existing customers and then amort these raises by increasing the Linux machines prices.
- Supplemental advertising
- etc.
So, if you want my 0.01$ guess, I'd say that the Linux laptop might be as expensive as the NT/2000 or at least ME ones.--
Re: (Score:2)
Re:Building a desktop OS from scratch (Score:3)
because OSs are layered things, and the lower levels (the kernel and drivers) are not anymore "designed to be a desktop OS" than designed to be anything else. The Linux kernel can be the foundation for a perfectly good destkop OS, it's just all the Unix command-line and /etc/configuration stuff that's historically tied to it. No-one's preventing YOU from making a desktop OS on top of the Linux kernel, that doesn't even use /bin/sh. For that matter, no one's preventing you from writing your own kernel either, but you'd better have some really good new ideas to put in it, to be worth the effort.
In the meantime, GNOME and KDE are moving forward, and doing the right thing: putting a pretty user interface on the whole system (which means that the average user ultimately doesn't have to touch the command line unless he actually wants to), while stayign reasonably close to the Unix way-of-doing-things, and keeping interoperability with more traditional Unix desktops (X11 + whichever wm).
Re: (Score:2)
Dell's think-tank (Score:5)
Dell: Nah, Linux sux
manager: yes, but it gets a lot of coverage lately. We should really come up with something.
Dell: Nah, Linux sux.
manager: Michael, you will slashdoted.
Dell: Darn, you are convicing. Call our PR department...
Building a desktop OS from scratch (Score:5)
This isn't to say that Linux is a bad OS. It's a terrific OS for servers, routers, and other non-end-user computers. But it doesn't make any sense to try to hack shiny, happy desktop features into it. Everything has its purpose right? Linux's purpose is to be the best server operating system available (whether or not it succeeds is your call), not to battle Microsoft.
Rather than that constantly remake Linux in order to compete with Windows, it would make a great deal more sense for the FSF to create a brand new operating system designed from the ground up to be a desktop OS. Not only would this OS include all the necessary components for a desktop OS (GUI support built in from the beginning, no CLI, journaling file system, plug-and-play devices, advanced multimedia support, etc.), it would eliminate all the problems seen in current desktops -- licensing problems with KDE; feature bloat with GNOME. And right now, there's simple no free OS that does this -- sure, there's BeOS, but it's only free as in beer, not as in speech.
Remember P.T. Barnum's famous quote "You can please all of the people some of the time, or some of the people all of the time?" Right now, Linux is trying to please all of the people all of the time, and that just isn't working. It's time to divide and conquer. Leave Linux to the server market and design the efficient, stable, user-friendly, and most importantly, open-source desktop OS that the world has been waiting for.
System administration vs typical usage (Score:2)
The problem is to set up the system. You must give them a preinstalled Linux and they must have someone who will fix problems for them. If they don't have a tech-savvy friend / neighbor / relative, they're screwed. And frankly, I don't think the people who need that much support (e.g. for every icon that disappeared) are able to solve the same problems under Windows.
Re:Building a desktop OS from scratch (Score:2)
I'm a desktop user, and I need its functionality.
There are a couple of different flavors of desktop user...
At one extreme, there's my mother, who expects her computer to boot up and work. She's not going to configure it, at least not intentionally. She wants a truly simple, iOpener-like system that lets her read and send email and see pictures of her granddaughter on the web. She uses a desktop all the time.
At the other are the sysadmins, developers, and IT folks who understand the hardware, the software, and how it's all glued together. They want X, ssh, and a wide selection of tools and capabilities; and if they can't find what they want, they'll build it themselves. They use a desktop when it's convenient.
Halfway in between are the "power users", the folks who know there's more to the machine than the desktop. They might be comfortable learning a bit about the shell, or a scripting language; but that's not their job. They want options and flexibility, but not too much - just enough to keep it all straight in their heads and get their jobs done. They use a desktop unless there's a very big advantage to using a CLI.
To be honest (and I don't mean to flame you), it sounds like you'd be happier using something like BeOS or MacOS. Why go to the trouble of using a complex, powerful Unix-like kernel if you're immediately going to rip out half of its functionality?
For a couple of reasons...
That last one is the real killer. I put "upgrade" in quotes because with the same system being used by multiple types of desktop users, it becomes a whole lot simpler for someone to learn a little bit here, a little bit there, and slowly expose the full functionality of a system by adding or exposing new pieces as they feel the need to.
Re:Not to put a downer on things (Score:2)
BeOS, Next, all very innovative. Amoeba, Minix, and loads of dusty academic OS's exist.
But nobody really uses them. I've seen seasoned developers weep with joy once they've played with the BeOS API, system admins go green with envy when they see the security & file system of Amoeba. But there is no takeup by industry.
Why is that?
One word : interoperability
Look at all the grief that is required just to get WinXX talking to Linux. Allison, of Samba fame, goes as far as saying "SMB sucks!". Yet we're still tied into the same ways of doing things. It's the profit motive for the S/W companies that actually rewards them for stopping cooperation between OS's (MS Kerberous, anyone?)... So innovations aren't easy to migrate to, it's an all or nothing thing in a lot of cases. Add to that the cost of retraining _all_ staff, and the costs are horrendous.
The only way I can see out of this is to get some Standards police, and have a unilateral legislative agreement that STANDARDS MUST BE ADHERED TO.
Then you get the companies whining that they cannot "innovate" for competitive advantage, and xth Amendment agitators (in the US) saying how the Government is censoring/stifling free expression.
In short, hte companies have us by the short and curlies and they won't let go until they're smacked down _hard_. I don't see this happening.
Strong data typing is for those with weak minds.
Re:Not to put a downer on things (Score:2)
This is probably the key to your question, by expecting an imaganary "open source" community to take your project of choice and blast it into the stratosphere, you are creating unreasonable expectations.
You are looking at the headline *NIX Open Source projects, so of course you can't find anything else. Most non *NIX platforms don't have an open source community, they have freeware, and shareware, but not free open source.
Nothing stops you from starting your own open source project. Just don't expect people to support you if you don't toe the party line.
PocketIFS [palmgear.com] is an open source fractal hacking tool for the PalmOS. It has receaved favorable reviews, one bug report, and not a single line of code from anyone but me. (Perhaps because it Just Works).
Thad
Dell's take on the NC... (Score:2)
Dell: No. As you get more bandwidth the computing power will expand both at the edge of the network and at the center.
Larry's model is that it expands at the center, but the edge gets less and less powerful. I think that actually both will get more and more powerful and as you get a faster connection to the internet, the data streams become more and more complex and there's richer and better data that gets processed and dealt with at the client. You can kind of see this a little bit if you talk to people who have DSL connections or cable modems in the home and you say to them, "Now that you've got this really fast connection, is your computer fast enough? Will you use a slower computer now that you have this really fast connection?"
Nobody wants to go to a slower computer now that they have a fast connection; they want a faster computer because the bottleneck shifts. Right now for a lot of users -- at least in small businesses and at home -- the bottleneck is the line, not the computing power, but when you get a fast connection, the bottleneck shifts back to the computer itself.
I agree with Dell that clients on the edge will grow in power just as clients in the center do. However, I don't think that's at all incompatible with a NC view of computers - NC's will be powerful devices, but more specialized.
For instance, I would love a small computer in the kitchen to look up recipies, and a small computer by the TV to be able to look up related things during TV programs. The kinds of devices I want in each place have a fairly different set of requirements though, which would best be served by an NC type of device. I could see such devices outnumbering normal computers in most households before too long, once easy household connectivity is figured out.
Also, now that I have a DSL line I do not find my computer (only a P450!) to be the bottleneck. I think my bandwith would have to increase by an order of magnitude or more to start thinking that.
What I found more interesting in the interview (Score:3)
From the interview:
Then I started going out on the Net and searching for "Dell and Linux," "Optiplex and Linux" "Dimension and Linux," "Inspiron and Linux," "Precision and Linux," and Boom! You'd see hundreds of thousands of user who had figured out how to use our products with Linux, and we hadn't done anything to help those people either.
In reality, though, (and this is something I find very, very surprising) many/most hardware companies are far slower than Michael Dell at catching on. Recently, I bought a logitech cordless mouse, but only after a frustrating search to find out if it works for linux. Now, logitech could hire a couple of web developers to put that info on their site, and maybe get a couple of their programmers to test stuff for linux. Bingo. You're helping out 20+ million customers you weren't before. Why don't CEOs see this?
My guess is that Michael Dell is probably way ahead of the curve, since he actually bothered looking at the search queries, installing linux himself, etc. Maybe most CEOs wait for a Gartner analyst to draw pie charts and wait for their underlings to draft a business plan before they get the point. Any idea how this works?
(As an aside, it's very revealing that Michael Dell actually saw *hundreds of thousands* of customers helping each other out before the company even realised there were all these customers using Dell on linux. Is it any surprise people prefer looking for product info from other users instead of the "official" company site? )
w/m
Re:Not to put a downer on things (Score:3)
The Free Software/Open Source community has plenty of choices. Some people develop for *BSD, many for Linux, a few for BeOS, and a very few brave and masochistic souls for Windows/*.
Most have chosen Linux for a variety of reasons, ranging from the licensing (GNU GPL) of the OS and underlying libraries to the simple fact that, for whatever reason, right or wrong, welike and prefer Linux. This may not reflect your preference, but it does reflect theres, and if you don't like it you are free to start your own project, or fork off any of the free software projects you wish.
The community hasn't ralleyed around Linux because it doesn't have a choice, its ralleyed around Linux because its license (GPL) and quality reflect the desires of the community more than anything else out there.
That having been said, there is a (smaller but just as enthusiastic) community which has formed around *BSD, also largely because they favor the license (FreeBSD) and the quality of the platform more than anything else out there.
BeOS is neat, but proprietary. Bug fixes are arduously slow in coming, precisely because the OS is proprietary and there is limited manpower fixing bugs, creating new features, etc. This makes the platform as a whole less dynamic and less satisfying for me, the end user, when I discover my video captures die after 9 minutes and don't get a fix for six months. It is hard to form a community around a product one cannot contribut bug fixes to, or get bug fixes from in a timely manner. while there is a small, vocal, and enthusiastic developer community or Be despite this, the proprietary nature of the platform (with all its disadvantages) make it singularly unattractive to open source developers regardless of how nice its programming API may be.
Ditto for MacOS. Proprietary OSes are simply unappealing to most open source developers.
Ditto for windows (with the caveat that the APIs are ugly, the system even less stable and much less elegant than any of the aforementioned OSes, etc.)
We have plenty of choice, and most of us have chosen, with FreeBSD and Linux the clear winners. However, the beauty of the free software movement (and open source) is that our decisions in no way force you to conform -- you are free to take the source code and develope it for whatevre program you like. You even have the right to bitch and moan because the rest of us don't share your platform preference. However, the rest of us have the right to ignore you, or even chortle with amusement, should we in turn find your choice to be as silly as you find ours.
It is called freedom, and it is a wonderful thing.
Re: (Score:2)
Re:Still not easy to find on the web site (Score:3)
- OEM licences for Windows are much cheaper than the licences you buy in a store, probably very close to the cost of an OEM licence from Red Hat (which is not exactly the cheapest distro anyways)
- It takes more time (and hence money) to set up Linux than it does to set up Windows. Flame away, my brethern, if you must, but 't is a fact, that is to say, 't is fact if you want to set up Linux as a desktop OS (which it was not exactly meant to be).
Having said that, I'd of course much prefer them not to install anything, be it Windows or Linux, because I'm one of those control freaks who'd rather keep everything in their own hands. I'd much rather put stuff on my harddisk in a way that makes sense to *me* (and probably no one else) than to dig through an entire 13.7 Gig HD trying to find out where Dell has put the sources to the kernel.Of course they see linux desktops in the future (Score:3)
Myself, I'd not mind as long as they charge less for a linux install than for a win98 install, and have options like order 1 year phone support plan or just a cheapbytes cd shipped with the box. Myself, I don't want or need linux support that way, i can get any info I need from the net. So please make it an option, Dell.
bash: ispell: command not found
Re:Linux on the desktop and Dell (Score:2)
--
Games = desktop success (Score:4)
The real trouble with pushing Linux onto the desktop market as quickly as possible is that the infrastructure required to develop games easily is still in the process of development. Why games? Because games are where mindshare is in the eyes of Joe Sixpack buying a PC for his kids. Sure, he may claim that it's for "educational purposes", but if it doesn't run games as well then he's not going to buy it.
Whilst the latest versions of XFree86 attempt to go further in what hardware is supported and what features can be used, there is still no unified framework for the kind of features required for games and other multimedia applications. Sure, there's OpenGL and now OpenAL, but these are very much a work in progress under Linux, and even then rely on the goodwill of manufacturers in making driver code available for a system where they won't be making any kind of return.
The reason that there are so many games under Windows is that it is, relatively speaking, easy to create them. DirectX provides a unified framework for integrating graphics, sound, music, input devices, network play and more in a single package, allowing designers to concentrate on what matters - the game itself. The difference is very real, and can be seen in the amount of time it takes for Loki to port a game to Linux.
It is safe to say that the importance of games cannot be stressed enough in the public's view of how desirable a desktop system is. And until Linux delivers a unified framework for creating games a la DirectX, it's success on the desktop will always be limited.
Slashdotting as a propaganda tool. (Score:2)
Dell: Nah, Linux sux
manager: yes, but it gets a lot of coverage lately. We should really come up with something.
Dell: Nah, Linux sux.
manager: Michael, you will [be?] slashdoted.
Dell: Darn, you are convicing. Call our PR department...
Interesting point...
Are there any Windows-related news media that have such a large following that mere mention of a URL will bring down a server? I don't THINK so... (Otherwise it would be called "WinNewsing" or some such rather than "slashdotting".)
So perhaps this can be turned to Linux's advantage...
"Look, guys. There are so many Linux fans on this one news board that just the mention of your site there will bring so many hits that you'll think you're under a DOS attack. Don't you think there might be a market for a linux port of your ?"
(Yes, I know there's some *BSD, BeOS, and Mac fans here too. But I still think it's a fair characterization.)
Re:Not to put a downer on things (Score:2)
Historically, this has come about because the tradition of opening source really began with Unix-like or pre-Unix systems on minicomputers. That whole community coverged on Unix, and the poeple who carried open source through the Dark Age of the 1980s were people from Unix backgrounds. Thus Sun used a variant of BSD, Gnu software is all *like* Unix software, and the only open source software for the early 16-bit home computers made them more like Unix machines.
Its not that we all think Unix is wonderful. It has many problems, the biggest one being X, but people are unlikely to abandon it for an alternative just for the hell of it. The Unix Way of doing various things has developed over quite a period of time, and has proved better than, or at least equal to, most other paradigms in OS design (most importantly the VMS and NT Way). A great deal of useful software, especially in the internet server arena, which is becoming of more concern even for small business and home users, is very much Unix-centric, as is seen in the difficulty of porting it to different OS models (Apache or Sendmail on NT don't work very well).
Now I'm going to start telling you what you really think, so please feel free to tell me I'm wrong
Most people who are looking for alternative to Unix, including the various Free Software projects that have been set up to persue this goal, are people who do not know a hell of a lot about OS kernel design, but who see the surface features of Unix - a slow, ugly, insufficiently standardised GUI, and a hard to understand CLI - and want a more coherent and simpler user experience.
My point is that there is no fundamental difficulty in creating these things on top of an otherwise Unix-like system. It needs some considerable thought, and it also probably needs a bunch of existing stuff, such as conventional CLIs, and X (at least in its current incarnation), and possibly even the model of IPC to be binned, or hidden from the user.
In a nutshell: not only is there no chance of getting the Open Source community to adopt a different OS, but there's no really good reason to ask it to. Much better to build on what we have. The much more important problem, in my view, is getting open source people to take on the concerns of "ordinary users".
Re:Building a desktop OS from scratch [OT?] (Score:2)
Leaving aside the fact that traditional tools have been used in art (not just to create art) and that an experienced professional craftsman does come to know "his" tools...
A physical hammer is akin to the original "word processors" (remember those beasts?) or a non-programmable calculator. Each is built for one task; you may be able to use it to do something else, but that's purely an accident of form and design. There's a reason why word processor machines are almost extinct - we (as a culture, possibly as a species) have decided that advantage of increased flexibility is worth the cost of increased complexity.
Computers are the first truly general purpose tool that mankind's invented. Imagine a chunk of matter that responded to your thoughts; if you think clearly enough, and in the right way, you can cause that chunk o' stuff to take the form of any tool you've ever seen, or any tool that you can imagine.
That's what a computer is. Not a hammer - but something that can be a hammer, or a screwdriver, or a socket wrench... as long as you know what you want, and how to turn it into the tool your interested in. The complexity is not in the use or the particular shape that the tool takes, but the fact that it can take many (any!) shape you can imagine, either perfectly or imprefectly suited for the job at hand.
Modern tools already do and will continue to incorporate elements of computers. That's progress for you. But don't confuse the technology that goes into the tool with the tool itself. Just because it's used in a simple tool doesn't mean that it's simple, or that more complex uses of that same technology are wrong.
Re:No M$ Tax ? (Score:2)
I bought it because all of its components were standard (and also because it was twice as less expensive as a Dell laptop)
- PIII600
- ATI Rage LT Pro
- Maestro 2E
- (and, BTW, a zone free) Toshiba DVD
- Tulip compatible Network card
I installed OpenLinux 2.4 on it and I have to say that this works.So, as I might expect most people here just spend more time choosing a laptop than a car, I am sure that having a dedicated Linux install for their laptop is not *that* crucial. BCNU
--
CNN (Score:3)
- the introduced the story with a big Linux logo,
- they pronounced it right^w the way I do,
- they were apparently filming at the show, though they actually focused on content rather than on the event,
- they used the de rigeur "scrappy upstart" description,
- they said it was popular because of its "amazing reliability" (sorry; forgot the exact words),
- they showed barricades labeled "Windows Free Zone",
- they showed systems running GUIs, while voice-over talking about Linux,
- they had a bite from an Eazel guy, mentioned their Appleness, and showed a bit of Nautilus,
- they said Linux was was used on 10% of servers and 4% of desktops,
- they showed an xterm (or the like) with an "uptime" display of 41 days (though I doubt that the uninitiated viewer would have been able to figure that out),
- they showed part of MD's speech (see, I really am on topic!).
CNN repeats some of these spots several times a day, so you may be able to catch it if you're inclined to.The most interesting thing about this is that it puts Linux in front of a very large audience, and they portrayed it in a very good light while doing so. You may have people asking about that Linux thingy you run, if they saw this.
--
Re:What do you want from a desktop OS? (Score:2)
-B