Hi folks, you say you want a new Linux car? Step right into our training room where we will go through our special course "How to buy and configure your new Linux car 101", and we will have you out and ready to actually select and buy your new car in just a matter of hours!
What? You think you want to purchase that Dodge Caravan at the dealer down the street? Hell no! You don't want that!! Why, don't you realize that our Linux vans can be configured with no less than 23 different kinds of seats? We will even give you a map to show you the way to the 23 different aftermarket Linux car parts shops that sell them!
OK, enough talk of other inferior cars! Come on in, take a seat, get comfortable. Let's go through the overview of what we will cover in the course..let's see..frames..yep, we will go through the 12 different kinds of frames and what associated parts and components you can actually buy in install on each. Engines.. great module.. we go over the 9 different engines you can buy, and also the many differrent carberator, cooling, and air conditioning systems available for each engine option, just terrific stuff.. what? yes, questions? Oh, you say you've really no mechanic experience? Well, don't worry! We provide a full set of Snap-On tools (of course, you could also choose Craftsman or one of our other 3 brands) and even let you borrow a car lift to make installation a breeze! Does that answer your question? Great! Now on to the next module, Suspension and Tire options...we'd better hurry if we are going to make it to the interior seating module by the end of day...can I get you folks some coffee?...hey! Where are you going?!? Come back! Don't buy that Dodge, they only have 3 models..come on, they don't give you the 38 differrent kinds of headlights that you can put into OUR babys! THEY EVEN FORCE YOU TO TAKE DELIVERY WITH THE STEREO SYSTEM ALREADY INSTALLED!! HEY, COME BACK!!!......
I have to echo some sentiments already posted about increasing complexity and the ability to control the environment to prevent crashes (of which there are all sorts of kinds). I started in programming 20+ years ago but over the last decade have mostly been involved with systems integration and consulting, and so have run the gamut from designing programs to ripping out my hair over trying to use them and the systems they are supposed to be running on.
Many different platforms have been mentioned in these posts, but I think there is a common facet to them all; the ones requireing limited hardware and software compatibility (and support fewer applications) are going to be more stable. It's easier to control the environment, and there are far fewer developers and customers to have to work with
Microsoft is in a position unlike other OS vendors; They are providing and supporting a family of OS's that are trying to handle *every* kind of user from a single home user to small businesses to fortune 50 firms with tens of thousands of users, and at the same time are striving for interoperability with other operating\file systems, compatibility with probably tens of thousands of peripherals, hundreds\thousands of PC platforms, etc, and that finally have a huge installed base of somewhat computer illiterate customers.. And, they of course have a huge installed base of older OS versions that they have to support. I think any vendor in their position, with today's overall environment, is not going to be able to do much better in terms of limiting crashes.
To the open source supporters - I think that if open source gets what it wants - reaching the masses - it's going to have the same problems the larger the number of customers it gains, the number of platforms\peripherals\software products open source will be asked to be perfectly compatible with will increase drastically, and those combinations will be "stressed" much more than they are now by the larger volumes of users. It seems like open source took a long time to support even some basic common peripherals in ways that typical consumers could deal with. If you took an open source platform like Linux and suddenly put it on the current platforms of several million end users with legacy systems, I would guess there would be widespread serious problems, and those users would have no one to go to for quick fixes. No "typical" user is going to be satisifed with being told that if they have a problem, they have to wait until a random user in similar circumstances with a *nix programming bent decides to poke their nose around and try to fix the problem in their spare time. That might work for a small core of very common platorms and products, but won't hold up to the thousands and thousands of hardware and software components out in the real world today. Which is why I think that while a LAMP server on a common platform might be extremely viable for businesses, the same OS\hardware setup might (probably would) fail miserably for the typical end user masses as it stands currently.
All that aside, no matter what platform you are discussing- as a user of these systems, I find the basic architecture of the PC platform very frustrating, and..ah.. crude considering all of the years of development there have been. Just look at even the basic hardware design - why are we still having to risk finger cuts to insert stubborn PCI boards into tight slots and having to try to screw the damn paddle boards in - why no boards in pluggable cartridges like a video game console, for instance? It's a small (stupid) point, but kind of shows that after all of these years, we still have a platform that has way too many individual component suppliers (hardware and software) and little standardization, quality, & error control. And I think that when OS vendors have to let all of these vendors play directly in the same sandbox, trouble is always going to happen.
It seems to me that we are never going to accept, en masse, complete proprietary hardware systems from sin
Slashdot Mirror
Hi folks, you say you want a new Linux car? Step right into our training room where we will go through our special course "How to buy and configure your new Linux car 101", and we will have you out and ready to actually select and buy your new car in just a matter of hours! What? You think you want to purchase that Dodge Caravan at the dealer down the street? Hell no! You don't want that!! Why, don't you realize that our Linux vans can be configured with no less than 23 different kinds of seats? We will even give you a map to show you the way to the 23 different aftermarket Linux car parts shops that sell them! OK, enough talk of other inferior cars! Come on in, take a seat, get comfortable. Let's go through the overview of what we will cover in the course..let's see..frames..yep, we will go through the 12 different kinds of frames and what associated parts and components you can actually buy in install on each. Engines.. great module .. we go over the 9 different engines you can buy, and also the many differrent carberator, cooling, and air conditioning systems available for each engine option, just terrific stuff.. what? yes, questions? Oh, you say you've really no mechanic experience? Well, don't worry! We provide a full set of Snap-On tools (of course, you could also choose Craftsman or one of our other 3 brands) and even let you borrow a car lift to make installation a breeze! Does that answer your question? Great! Now on to the next module, Suspension and Tire options...we'd better hurry if we are going to make it to the interior seating module by the end of day...can I get you folks some coffee?...hey! Where are you going?!? Come back! Don't buy that Dodge, they only have 3 models..come on, they don't give you the 38 differrent kinds of headlights that you can put into OUR babys! THEY EVEN FORCE YOU TO TAKE DELIVERY WITH THE STEREO SYSTEM ALREADY INSTALLED!! HEY, COME BACK!!!......
I have to echo some sentiments already posted about increasing complexity and the ability to control the environment to prevent crashes (of which there are all sorts of kinds). I started in programming 20+ years ago but over the last decade have mostly been involved with systems integration and consulting, and so have run the gamut from designing programs to ripping out my hair over trying to use them and the systems they are supposed to be running on. Many different platforms have been mentioned in these posts, but I think there is a common facet to them all; the ones requireing limited hardware and software compatibility (and support fewer applications) are going to be more stable. It's easier to control the environment, and there are far fewer developers and customers to have to work with Microsoft is in a position unlike other OS vendors; They are providing and supporting a family of OS's that are trying to handle *every* kind of user from a single home user to small businesses to fortune 50 firms with tens of thousands of users, and at the same time are striving for interoperability with other operating\file systems, compatibility with probably tens of thousands of peripherals, hundreds\thousands of PC platforms, etc, and that finally have a huge installed base of somewhat computer illiterate customers.. And, they of course have a huge installed base of older OS versions that they have to support. I think any vendor in their position, with today's overall environment, is not going to be able to do much better in terms of limiting crashes. To the open source supporters - I think that if open source gets what it wants - reaching the masses - it's going to have the same problems the larger the number of customers it gains, the number of platforms\peripherals\software products open source will be asked to be perfectly compatible with will increase drastically, and those combinations will be "stressed" much more than they are now by the larger volumes of users. It seems like open source took a long time to support even some basic common peripherals in ways that typical consumers could deal with. If you took an open source platform like Linux and suddenly put it on the current platforms of several million end users with legacy systems, I would guess there would be widespread serious problems, and those users would have no one to go to for quick fixes. No "typical" user is going to be satisifed with being told that if they have a problem, they have to wait until a random user in similar circumstances with a *nix programming bent decides to poke their nose around and try to fix the problem in their spare time. That might work for a small core of very common platorms and products, but won't hold up to the thousands and thousands of hardware and software components out in the real world today. Which is why I think that while a LAMP server on a common platform might be extremely viable for businesses, the same OS\hardware setup might (probably would) fail miserably for the typical end user masses as it stands currently. All that aside, no matter what platform you are discussing- as a user of these systems, I find the basic architecture of the PC platform very frustrating, and ..ah .. crude considering all of the years of development there have been. Just look at even the basic hardware design - why are we still having to risk finger cuts to insert stubborn PCI boards into tight slots and having to try to screw the damn paddle boards in - why no boards in pluggable cartridges like a video game console, for instance? It's a small (stupid) point, but kind of shows that after all of these years, we still have a platform that has way too many individual component suppliers (hardware and software) and little standardization, quality, & error control. And I think that when OS vendors have to let all of these vendors play directly in the same sandbox, trouble is always going to happen.
It seems to me that we are never going to accept, en masse, complete proprietary hardware systems from sin