If a professional music site sold high quality mp3's for a fifty cents to a dollar a pop from a fast, dedicated site then people would flock to it. Better yet if it had band news, discussion forums, reviews, ratings, live streaming etc. then people would live on it.
Naturally you'd still have lamers on p2p, but then these people would never use a pay site anyway, even if meant wasting ten hours to find and download the same songs that the pay site sold for seven dollars.
X11 still sucks big time. It's bloated, the API sucks, the configuration is arcane, font support is arcane, pointer device & monitor configuration sucks, the drivers are poor, you need widgets and a window manager to get it be half usable etc. It's not hard to see why Linux desktops looks so odd and behave so poorly compared to XP or OS X. Remote networking is great, but Linux is being used more and more and more locally, so perhaps it is time to consider a lighter more modern replacement.
I'm being serious here. When apps are run through abstract layers like GTK and QT, it doesn't matter whether X11 is underneath or not, so why not write something that runs faster locally, has a modern multi-threaded event model, better font support, display postscript like qualities? It could add all kinds of cool features such as mapping windows into hardware textures etc. that the likes of Aqua enjoys.
And what about the people who want to run remote apps? Well they can run X11 as they already do or do what X11 does on OS X - run in rootless mode on top of the native WM. Apps would run either way so long as implementations of QT & GTK had the same binary interface.
Not to mention that Walmart already sell a bonafide PC already running Linux complete with 128Mb and 20Gb harddrive for $249. Not only is a proper PC, but it's faster and more useful too, seeing has it a keyboard, mouse, builtin ethernet, a modem and a array of standard ports to plug things into. It would make a perfect firewall or mail server.
Frankly there is little point getting an XBox except as a toy, or unless you have a vested interested in breaking the encryption to sell games without a Microsoft tax.
The biggest scam is where these guys will charge you $9.95 or more for a personal email certificate which expires after a year and if you read the small print they do not guarantee the authenticity of the sender in any way whatsoever. Tell me why then that someone wishing get a cert has to jump through all kinds of hoops such as supplying their social security / passport number etc. and fork out money for what is essentially a worthless certificate? You might as well generate a PGP key for the same level of authentication it offers.
I can appreciate that companies might find value in the corporate levels certs which bestow a certain degree of authenticity but certainly not email certs.
Someone such as gnu.org, or another free software group such mozilla.org should set up a free server that dishes these things out on demand - sign up, wait for the confirmation email, click on the unique link and get your cert. It's certainly no less secure than what the likes of Verisign would sell you and it would dramatically increase the use of encrypted and signed email. This would be a good thing by any measure.
Sorry, but Mozilla does a wonderful job on the internet. It copes with bad pages extremely well and I agree that it would suck if it didn't, but fortunately a lot of effort has gone into quirks mode.
I say that as someone who uses it all day with rarely an issue. I do run across the rare site that doesn't work, such as those relying on ActiveX, but the biggest problem is sites that look at the user agent string and point blank refuse to proceed.
Mozilla is an excellent choice for deployment in large organizations. It is possible to customize it in numerous ways, is easy to deploy, runs unchanged on various versions of Win32, doesn't replace system DLLs and is *much* less of a security headache than IE / Outlook (Express). Personally if I were an admin I would be sorely tempted to dump IE and Outlook altogether on this last point alone.
Re:Why can't they arleady do this?
on
SVG On the Rise
·
· Score: 2, Insightful
I don't see what would stop Mozilla using an LGPL library. Even proprietary software can use LGPL libs so long as any changes to the lib itself are made public. And Mozilla certainly isn't proprietary seeing as it's mostly tri-licenced as MPL/LGPL/GPL itself. As all of the source of Mozilla is available via CVS anyone else can grab changes quite easily.
Browsers that render bad sites badly are bad browsers. Like it or not, HTML from 4.0 downwards was never properly defined, never properly rendered (by browsers), never properly generated (by editors, perl scripts) and never properly written by human beings. If your browser doesn't handle these quirks then a fair proportion of the web will not render properly. Try explaining this to your users and they'll say "well it works in IE/Netscape/Mozilla/Opera...". Try chasing up N (where N > 1000) websites (good luck advocating your case to the Chinese, Urdu, Finnish etc. site admin) that your users want to get at but which don't render because there is no DOCTYPE or because the site puts the wrong tag inside another tag, or where the JS document.writes an infinite number of nested IFRAMEs. You can't win by not supporting these sites and Apple users must rank alongside AOL users as being the least clueful and least likely to understand why you won't just fix your browser.
I agree there is a case for not rendering XHTML properly or other well-formed content but while HTML exists browsers have got to grin and render it even if it does taste like shit.
And all of the above have visible menus and buttons that tie in to.mac, to buy CDs / airline tickets / hotels / movie tickets through Apple partners, to listen to their radio stations, to sync with an iPod etc. So they use open standards to direct people to their proprietary services - so what?
What makes a browser is not how well it renders good sites (which of course it should do anyway) but how well it renders bad sites. Huge swathes of the web are made up of gnarly, shitty, broken HTML, frame abuse, CSS, images and Javascript. Browsers that balk at that are bad browsers irrespective of their code size.
Now to concentrate on Chimera (since Mozilla also includes mail/news clients, HTML editor, JS debugger, DOM inspector, Cookie manager IRC etc.). Is is slower? Not noticeably as far as I can tell (I'm using it right now) and it runs fine for me on my crappy 450Mhz Mac. Is it less Mac-ish? Nope, in fact Chimera is probably more compliant with UI guidelines than brushed metal Safari. Could it be made smaller? Probably yes since so far Chimera has brought its own socket code, portable runtime library, image decoders, network decoders etc.. so at least some of these could be dumped in favour of the system equivalents (though it might impact stability or performance).
So aside from the hurdle of download size, what matters at the end of the day is which is a better browser. Apple had better put out a browser which has a decent browsing experience or they're going to be clobbered. Both browser engines will improve over time, but IMHO Safari has a long way to go yet before it is remotely comparable in terms of sheer quality or stability.
They eschew standards much better than they used to but many of features in the iApps (iDisk, iCal, iTunes etc.) and.mac initiative still smack of proprietary lock-in in one form or another.
This guy has reeled out one negative, uninformed article after another about Mozilla. This article is no exception - the guy just cuts and pastes the most salacious remarks out of context to spin a negative picture on any situation. I guess in a way it is not surprising he's now dredging weblogs for quotes since no one associated with the project would give this hack the time of day.
It is a toss up between him and MozillaQuest for who can spout the biggest load of bullshit about the browser. Personally I wonder what the hell Mozilla or Netscape did to these guys in the first place that they've carried such a chip on their shoulder about the project ever since.
So are there many other people in your area who wear that exact same combination of clothes and shop at the same store?
At worst (for you) it could know exactly who you are especially if you bought all those things in that chain store. At best it would still let them know your sex (unless you're a cross dresser), and can make a good guess of your age, lifestyle, weight and dimensions.
Now imagine a few scanners set strategically around the store and at the cash register and that you take a look around and then purchase another item of clothing with a credit card (assume you payed cash before).
Before you walked in they knew nothing about you. Now they know your name, your credit card number, the clothes you're wearing, how long you've been in the store, what part of the store you looked at most, what part you skipped, your approximate weight, lifestyle and age. And all you've done is buy a pair of socks!
Would any store go to the effort? Probably not until the technology improved, but I wouldn't put it past them. In fact, I can imagine that store cards of the future would employ similar technology so that the moment you walked in the door carrying the card they'd know who you were.
Apple products don't exactly work for humans either unless you're a clueless newbie. OS X is an example of an OS which is wonderful to behold and great for newbies but can be an enormous pain in the backside if you have to sit in front of it the whole day, having to put up with a lobotomized interface, poor online help and some grotequely screen wasting UIs.
Well you could always use the on-screen keyboard or invest in a real keyboard. Personally I like Grafitti although it does get a bit confused sometimes especially if you write in a 'slight' manner or are drunk at the time.
This story sounds like utter, complete nonsense to me. Simply put, it would be absolutely impossible for one person to write an acceptably working browser let alone one that can play DVDs, 4x surfing etc. in that space of time. I strongly suspect the chap has filched a bunch of code from various other projects (e.g. Mozilla, Xine), lashed them together and claimed them as his own. Even worse would be if it transpired he just glued a bunch of ActiveX controls together with VB onto a form.
As for his speed claims, this too sounds like hype. I can think of several ways to 'speed up' surfing, (e.g. trickle precaching sites, tinkering with QoS settings etc.) but all of them are just making better use of existing bandwidth and browsing habits and certainly wouldn't increase speeds by the multiples he suggests.
I do not believe for a second he has done anything innovative at all.
I disagree. OS X software tends to have a simpler UI but that doesn't necessarily equate to higher productivity. If you're a power user you end up wasting a lot of time hunting for the advanced features which either a) don't exist b) are buried in a subdialog somewhere c) exist if you happen to know some obscure key sequence. Even apps like the finder have some appallingly stupid and annoying behaviour, and just feel clunky compared to explorer.
Overall my experience (and I own a Power Mac using OS X and a PC) is that the PC is easier. If I were a newbie then perhaps it would be the other way around, but since I'm not the blanket statement that OS X is easier is not true. That's not to say I don't use the Mac sometimes for convenience, but I don't see anything intrisically easier about it than the PC.
And that extra money would mostly go towards content side work that AOL do. I don't particularly like the AOL client software, but the content is excellent and extremely simple to navigate. I wouldn't be tempted to use it for shopping, but it must rank as the most simple convenient site if you want to grab a recipe, learn how to fix a crack in wall plaster, read film news or something of that nature.
I don't know about others but I find it an immense pain in the backside to debug anything on Linux. The reason is GDB which has to be the biggest hinderance to development on Linux. Arcane, slow and unintuitive would be words best used to describe it. You could grow a beard while waiting for a backtrace when debugging a stack overflow in an app like Mozilla. I've seen GDB page my system to death before now, such must its memory requirements must be for some operations.
Even front end tools are little help. DDD is much better than the cli but it still suffers the same performance faults as GDB and introduces some fun issues of its own. Even a 'commercial' tool such as Project Builder on Mac OS X groans to a halt with GDB running underneath and has some very odd concepts concerning when breakpoints are hit or not.
All this would be understandable if Win32 were the same but it isn't. Debugging on Win32 is a snap - just in time debugging, compile in time debugging, a debugger that works, integration with the editor and more. It is sad to say that if asked and irrespective of open / closed source issues, which OS were the better for development I would say Windows. In fact, if I'm faced with a bug in Mozilla, I'd rather fire up Windows and replicate the problem there than wade through the shit that GDB throws in my face.
Die GDB! Linux really needs a decent debugger and all the mod cons that Windows developers have benefited from in the past decade.
Eh? Gecko is threadsafe, assuming you use it correctly, calling objects on the same thread that you were handed an XPCOM pointer to them on. And if you want to call an object from some other thread you can do so assuming you marshal the call or know the object in question is free threaded. Most of this is transparent since XPConnect can create proxy objects that marshal calls between threads over an event queue.
In fact, it would prove quite a good fit for Cocoa, which isn't multithreaded either unless you explicitly kick off a new thread. And if you do go multithreaded then you could still call objects in the ways listed above.
Certainly Gecko may have some threading issues, but these are more to do with what-if scenarios that no one has tested, e.g. what if you run the layout engine on a worker thread? etc. My guess would be yes it would work but until someone tries, who knows?
As for KHTML, I doubt the answer is likely to be any different. A glance through lxr.kde.org suggests that large chunks of it are thread unsafe too, e.g. two threads programatically altering the DOM are not protected from stomping over each other.
Gecko doesn't fake the look of Aqua, it uses the Apple's own rendering engine to draw the buttons exactly as they should look. So if a page contains a checkbox field, Gecko will tell Aqua to render the checkbox at x, y, width, height with the style appropriate for its checked / unchecked / depressed / enabled / disable state and so on. I haven't looked at QT but I would not be surprised if did exactly the same. If it's not doing exactly the same, then Safari is going to have all sorts of fun ordering issues when it encounters forms in DIV elements and so forth. Since Chimera has a native UI, both browsers are the same with regards to the trimmings around the engine.
CSS also allows form elements to be stylized. Using native widgets makes that nigh on impossible to support.
In terms of performance, my experience is that a good browser is one which handles the best and the worst of the web equally well. Konq / KHTML always struck me as one that coped well with the middle content but took a nosedive with anything else. I've downloaded Safari now to play with so I might revise my opinion with more exposure.
This choice sounds utterly insane to me. With the greatest respect, khtml is nowhere near as good as Gecko in terms of it standards support or behaviour or stability especially when dealing with some of the crap sites out there in the world. Run it through a few random sites involving nested tables, CSS or frames and it quickly screws up rendering.
What the hell were they thinking? Perhaps it's a little faster or smaller, but that sounds like a small payoff when you end up with a browser that is broken and doesn't work properly on a large number of sites. Chimera shows that Gecko can make an amazing browser on OS X so why they've jumped over is mind boggling.
It is victimhood, but there is a valid point here which you seem to have completely missed or ignored. It takes hundreds of hours of character development before you start to see the faults, but by that point there is a large incentive to keep playing - friends you've met, the character you've spent so long building up, the 6 month sub you have left etc.
Of course people could just stop (the same as with any addiction), but when you reach that point your perspective is skewed. You don't want to give up the time you have invested and will overlook the faults hoping against hope that Verant will get off their arses and fix the problems with the next expansion or update. Of course what usually happens is the problems don't get fixed, but the cycle repeats.
I was in the same situation myself. Although I wasn't "hardcore" (playing an hour or two at most a night), and have a real life too, I found it quite a struggle to leave. I can well imagine that others would find it nigh on impossible. Ironically for me it was Verant themselves who managed to snap me out it. The Shadows of Luclin expansion release was such a hamfisted, buggy travesty that I lost all confidence that they ever gave a crap about the game. Considering the ludicrous amounts of money they made each months on subs I was damned if I was going to put up with it any more.
It doesn't sound from their pre-release attitude with their Star Wars rpg that they have learnt from the experience. SW players can look forward to being fucked over and ignored, bugs will go unfixed, the game will be repetitive, and expansions will still be buggy, but a hardcore will still be addicted despite that.
Mandrake isn't the most user friendly distribution. Not by a long stretch.
It's desktop is a stock KDE cobbled together with homegrown tools which border on the shoddy. Aside from inheriting the clutter of a generic KDE, it's littered with its own usability problems and in general is a mess. It's as if they let their programmers put any shit they like into the UI without subjecting it to QA or usability testing.
Red Hat is miles better and I say that as someone who switched from Red Hat at 6.2 to Mandrake 7.0, put up with progressively worse releases until 8.3 when I made the jump back again. I did so because I saw no clue whatsoever that Mandrake even gave a damn about usability. And this is funny considering how they're trying to get into the preinstall business. If Mandrake wants to win more users it is going absolutely the wrong way about it.
The UI in Red Hat 8.0 is extremely impressive (for Linux). Perhaps with a few more point releases it will be on par with Windows or OS X. It certainly beats Mandrake to a bloody pulp in terms of usability and on several other fronts. I do miss the large number of packages that Mandrake has, but I consider the price small when they can be fetched by hand if need be.
Imagine the Dell customisation page was like this:
Choose your operating system
Windows XP Home
Windows XP Professional (add $150)
Red Hat Linux 8.0 Personal (deduct $100)
Red Hat Linux Professional
No operating system (deduct $150)
Given that, do you really think Dell would install Windows on 99.9% of PCs? I don't. In fact I believe the figure would be substantially less.
And not just amongst Linux fans either, since lots of people already have a copy Windows floating around. Whether its 'legal' or not is immaterial as far as Dell should be concerned since that's a matter for the customer.
Does it really replace them or install them somewhere else? Fink shoves everything into a/sw/ folder so it won't overwrite anything. It won't even be picked up by the system by accident since you must explicitly set paths by sourcing a shell script in your.profile to set it up.
It still requires you run sudo to install it however, though presumably you might be able to make it install with lesser privileges if you did some chown ground work on/sw first.
Slashdot Mirror
Naturally you'd still have lamers on p2p, but then these people would never use a pay site anyway, even if meant wasting ten hours to find and download the same songs that the pay site sold for seven dollars.
I'm being serious here. When apps are run through abstract layers like GTK and QT, it doesn't matter whether X11 is underneath or not, so why not write something that runs faster locally, has a modern multi-threaded event model, better font support, display postscript like qualities? It could add all kinds of cool features such as mapping windows into hardware textures etc. that the likes of Aqua enjoys.
And what about the people who want to run remote apps? Well they can run X11 as they already do or do what X11 does on OS X - run in rootless mode on top of the native WM. Apps would run either way so long as implementations of QT & GTK had the same binary interface.
Frankly there is little point getting an XBox except as a toy, or unless you have a vested interested in breaking the encryption to sell games without a Microsoft tax.
I can appreciate that companies might find value in the corporate levels certs which bestow a certain degree of authenticity but certainly not email certs.
Someone such as gnu.org, or another free software group such mozilla.org should set up a free server that dishes these things out on demand - sign up, wait for the confirmation email, click on the unique link and get your cert. It's certainly no less secure than what the likes of Verisign would sell you and it would dramatically increase the use of encrypted and signed email. This would be a good thing by any measure.
I say that as someone who uses it all day with rarely an issue. I do run across the rare site that doesn't work, such as those relying on ActiveX, but the biggest problem is sites that look at the user agent string and point blank refuse to proceed.
Mozilla is an excellent choice for deployment in large organizations. It is possible to customize it in numerous ways, is easy to deploy, runs unchanged on various versions of Win32, doesn't replace system DLLs and is *much* less of a security headache than IE / Outlook (Express). Personally if I were an admin I would be sorely tempted to dump IE and Outlook altogether on this last point alone.
I don't see what would stop Mozilla using an LGPL library. Even proprietary software can use LGPL libs so long as any changes to the lib itself are made public. And Mozilla certainly isn't proprietary seeing as it's mostly tri-licenced as MPL/LGPL/GPL itself. As all of the source of Mozilla is available via CVS anyone else can grab changes quite easily.
I agree there is a case for not rendering XHTML properly or other well-formed content but while HTML exists browsers have got to grin and render it even if it does taste like shit.
And all of the above have visible menus and buttons that tie in to .mac, to buy CDs / airline tickets / hotels / movie tickets through Apple partners, to listen to their radio stations, to sync with an iPod etc. So they use open standards to direct people to their proprietary services - so what?
What makes a browser is not how well it renders good sites (which of course it should do anyway) but how well it renders bad sites. Huge swathes of the web are made up of gnarly, shitty, broken HTML, frame abuse, CSS, images and Javascript. Browsers that balk at that are bad browsers irrespective of their code size.
Now to concentrate on Chimera (since Mozilla also includes mail/news clients, HTML editor, JS debugger, DOM inspector, Cookie manager IRC etc.). Is is slower? Not noticeably as far as I can tell (I'm using it right now) and it runs fine for me on my crappy 450Mhz Mac. Is it less Mac-ish? Nope, in fact Chimera is probably more compliant with UI guidelines than brushed metal Safari. Could it be made smaller? Probably yes since so far Chimera has brought its own socket code, portable runtime library, image decoders, network decoders etc.. so at least some of these could be dumped in favour of the system equivalents (though it might impact stability or performance).
So aside from the hurdle of download size, what matters at the end of the day is which is a better browser. Apple had better put out a browser which has a decent browsing experience or they're going to be clobbered. Both browser engines will improve over time, but IMHO Safari has a long way to go yet before it is remotely comparable in terms of sheer quality or stability.
They eschew standards much better than they used to but many of features in the iApps (iDisk, iCal, iTunes etc.) and .mac initiative still smack of proprietary lock-in in one form or another.
It is a toss up between him and MozillaQuest for who can spout the biggest load of bullshit about the browser. Personally I wonder what the hell Mozilla or Netscape did to these guys in the first place that they've carried such a chip on their shoulder about the project ever since.
At worst (for you) it could know exactly who you are especially if you bought all those things in that chain store. At best it would still let them know your sex (unless you're a cross dresser), and can make a good guess of your age, lifestyle, weight and dimensions.
Now imagine a few scanners set strategically around the store and at the cash register and that you take a look around and then purchase another item of clothing with a credit card (assume you payed cash before).
Before you walked in they knew nothing about you. Now they know your name, your credit card number, the clothes you're wearing, how long you've been in the store, what part of the store you looked at most, what part you skipped, your approximate weight, lifestyle and age. And all you've done is buy a pair of socks!
Would any store go to the effort? Probably not until the technology improved, but I wouldn't put it past them. In fact, I can imagine that store cards of the future would employ similar technology so that the moment you walked in the door carrying the card they'd know who you were.
Apple products don't exactly work for humans either unless you're a clueless newbie. OS X is an example of an OS which is wonderful to behold and great for newbies but can be an enormous pain in the backside if you have to sit in front of it the whole day, having to put up with a lobotomized interface, poor online help and some grotequely screen wasting UIs.
Well you could always use the on-screen keyboard or invest in a real keyboard. Personally I like Grafitti although it does get a bit confused sometimes especially if you write in a 'slight' manner or are drunk at the time.
As for his speed claims, this too sounds like hype. I can think of several ways to 'speed up' surfing, (e.g. trickle precaching sites, tinkering with QoS settings etc.) but all of them are just making better use of existing bandwidth and browsing habits and certainly wouldn't increase speeds by the multiples he suggests.
I do not believe for a second he has done anything innovative at all.
Overall my experience (and I own a Power Mac using OS X and a PC) is that the PC is easier. If I were a newbie then perhaps it would be the other way around, but since I'm not the blanket statement that OS X is easier is not true. That's not to say I don't use the Mac sometimes for convenience, but I don't see anything intrisically easier about it than the PC.
And that extra money would mostly go towards content side work that AOL do. I don't particularly like the AOL client software, but the content is excellent and extremely simple to navigate. I wouldn't be tempted to use it for shopping, but it must rank as the most simple convenient site if you want to grab a recipe, learn how to fix a crack in wall plaster, read film news or something of that nature.
Even front end tools are little help. DDD is much better than the cli but it still suffers the same performance faults as GDB and introduces some fun issues of its own. Even a 'commercial' tool such as Project Builder on Mac OS X groans to a halt with GDB running underneath and has some very odd concepts concerning when breakpoints are hit or not.
All this would be understandable if Win32 were the same but it isn't. Debugging on Win32 is a snap - just in time debugging, compile in time debugging, a debugger that works, integration with the editor and more. It is sad to say that if asked and irrespective of open / closed source issues, which OS were the better for development I would say Windows. In fact, if I'm faced with a bug in Mozilla, I'd rather fire up Windows and replicate the problem there than wade through the shit that GDB throws in my face.
Die GDB! Linux really needs a decent debugger and all the mod cons that Windows developers have benefited from in the past decade.
In fact, it would prove quite a good fit for Cocoa, which isn't multithreaded either unless you explicitly kick off a new thread. And if you do go multithreaded then you could still call objects in the ways listed above.
Certainly Gecko may have some threading issues, but these are more to do with what-if scenarios that no one has tested, e.g. what if you run the layout engine on a worker thread? etc. My guess would be yes it would work but until someone tries, who knows?
As for KHTML, I doubt the answer is likely to be any different. A glance through lxr.kde.org suggests that large chunks of it are thread unsafe too, e.g. two threads programatically altering the DOM are not protected from stomping over each other.
CSS also allows form elements to be stylized. Using native widgets makes that nigh on impossible to support.
In terms of performance, my experience is that a good browser is one which handles the best and the worst of the web equally well. Konq / KHTML always struck me as one that coped well with the middle content but took a nosedive with anything else. I've downloaded Safari now to play with so I might revise my opinion with more exposure.
What the hell were they thinking? Perhaps it's a little faster or smaller, but that sounds like a small payoff when you end up with a browser that is broken and doesn't work properly on a large number of sites. Chimera shows that Gecko can make an amazing browser on OS X so why they've jumped over is mind boggling.
Of course people could just stop (the same as with any addiction), but when you reach that point your perspective is skewed. You don't want to give up the time you have invested and will overlook the faults hoping against hope that Verant will get off their arses and fix the problems with the next expansion or update. Of course what usually happens is the problems don't get fixed, but the cycle repeats.
I was in the same situation myself. Although I wasn't "hardcore" (playing an hour or two at most a night), and have a real life too, I found it quite a struggle to leave. I can well imagine that others would find it nigh on impossible. Ironically for me it was Verant themselves who managed to snap me out it. The Shadows of Luclin expansion release was such a hamfisted, buggy travesty that I lost all confidence that they ever gave a crap about the game. Considering the ludicrous amounts of money they made each months on subs I was damned if I was going to put up with it any more.
It doesn't sound from their pre-release attitude with their Star Wars rpg that they have learnt from the experience. SW players can look forward to being fucked over and ignored, bugs will go unfixed, the game will be repetitive, and expansions will still be buggy, but a hardcore will still be addicted despite that.
It's desktop is a stock KDE cobbled together with homegrown tools which border on the shoddy. Aside from inheriting the clutter of a generic KDE, it's littered with its own usability problems and in general is a mess. It's as if they let their programmers put any shit they like into the UI without subjecting it to QA or usability testing.
Red Hat is miles better and I say that as someone who switched from Red Hat at 6.2 to Mandrake 7.0, put up with progressively worse releases until 8.3 when I made the jump back again. I did so because I saw no clue whatsoever that Mandrake even gave a damn about usability. And this is funny considering how they're trying to get into the preinstall business. If Mandrake wants to win more users it is going absolutely the wrong way about it.
The UI in Red Hat 8.0 is extremely impressive (for Linux). Perhaps with a few more point releases it will be on par with Windows or OS X. It certainly beats Mandrake to a bloody pulp in terms of usability and on several other fronts. I do miss the large number of packages that Mandrake has, but I consider the price small when they can be fetched by hand if need be.
Choose your operating system
Given that, do you really think Dell would install Windows on 99.9% of PCs? I don't. In fact I believe the figure would be substantially less.
And not just amongst Linux fans either, since lots of people already have a copy Windows floating around. Whether its 'legal' or not is immaterial as far as Dell should be concerned since that's a matter for the customer.
It still requires you run sudo to install it however, though presumably you might be able to make it install with lesser privileges if you did some chown ground work on