Companies should be bound by law to maintain DTDs and XML converters for their proprietary formats.
Most proprietary file formats are by nature binary though, how would you represent say an image as XML? You can't of course, and often files are mixed textual/binary such as MS Word files.
Really I think there is a line to be drawn here - sometimes (as in the case with MS) file formats are being kept secret to facilitate lockin. This is clearly unacceptable, as it impedes the economy and restricts competition. I seriously doubt the MS Word file format contains vast amounts of innovation that everyone could steal if they were forced to document it.
On the other hand, especially with codecs for instance, there COULD be a lot of intellectual property involved that has cost time and money to research. Should they be forced to open up as well? Clearly not. Really this discussion is about, where to find the balance.
Really I think before machine translation is a workable reality, we'll need something like Cyc. To get a real, decent translation simple rule based systems aren't sufficient - you need to be able to translate the languages into logical formulae that can then be reconstructed into the new language. Cyc already has this ability to a limited extent (though only for english), but really, I think this is the way to go.
Indeed, I read somewhere that web hosters were loving the idea - "So you're telling me you get Apache on a box with built in ethernet and a 10gig disk, with Linux for $200, and it costs that little because BILL GATES PAYS FOR HALF OF IT?!?! Let me at 'em!"
Let's face it, who could resist the idea of getting a cool computer while at the same time losing Microsoft money? It's a fab idea!
We should be supporting the game companies that port to Linux instead of trying to get games working at 50% of the performance of Windows.
You're right, but unfortunately techniques for writing cross platform games are currently immature. SDL is about as good as it gets, and it isn't yet up to the standard of DirectX. Also, until OpenGL gets significantly ahead of Direct3D, games companies will continue to write to D3D, and therefore card manufacturers will write better Direct3D drivers (or no opengl drivers) as that's where the money is.
Steps to helping developers write cross platform games:
Improve SDL, bringing it up to the level of DirectX - this includes improved documentation. It might need professional support in the form of a company too.
Create copy protection systems that work on Linux, or convince existing companies to port theirs. Yes yes, I know that will upset a lot of people, but tough. The games companies charge for their products and want them protected, if that requires kernel patches then so be it.
Evangelism and support! Buy only nVidia cards, as they a) produce excellent cards and b) write good OpenGL drivers unlike say (cough) Matrox. My current card is a G400 - my next will be an nVidia card. Then I'll send them a polite email thanking them for their efforts and support. It can only help.
Ditto for games companies. Evangelism - write to SDL, it's easier, faster AND it'll make cross platform code.
Read the whitepaper - GNOME 2 support is in there, and in the PDF is says actually only KDE Base is mandantory. Some of the details seem a little inconsistent right now, I think they're still figuring out exactly what it'll be.
2) Greatly reduce the storage space consumed by hosting RPMs. Instead of needing storage space for four separate complete sets of RPMs, only one set of RPMs need to be hosted. These will be guaranteed to work on all four distributions.
Actually according to the white paper, there will be RedHat compatible libs/symlinks,so in theory 1 RPM will work on 5 distros. I'm not sure I like this however, as it'd effectively make RedHat the "standard" to which RPMs would be targeted, as then you know it'll work natively on redhat, and with the compatability mode on UL systems.
Hmm, I'm pretty sure ALSA is included in most distros these days. SuSE 8 I know has ALSA 0.9 in it, which is pretty good. I'm using ALSA now in fact and I'm on 7.3
Just because the official kernel doesn't have something doesn't mean it won't be there remember.
Unfortunately there are some distros that simply cannot face the large effort required to make themselves LSB compatible. For instance the LSB says you must use RPM - how likely is it that Debian will listen to that and drop apt/dpkg: it's primary selling point?
RedHat derived distros like Mandrake are also seeing problems with stuff like file paths and so on.
Car analogies are often flawed - remember that cars already ARE standardised in that they all are roughly the same width, height etc. They all have similar turning circles: basically they all work on our roads in the same way. In computers we have a situation where some cars can only go down some roads and it's a mess. Roll on LSB.
Important KMail fix
on
KDE 3.0.1 Ships
·
· Score: 5, Informative
# kmail: Don't eat in some cases the whole folder, when moving messages between IMAP folders.
This is one that I noticed on the KMail homepage. If you use KMail with IMAP then please please upgrade for your own sake, or else massive dataloss could occur! I'm not sure how the KMail team let this one slip past, but it's a pretty important fix.
KHTML and Gecko serve two different purposes actually.
KHTML = small, light, fast, good at displaying HTML quickly. Ideal for embedding, which is what it's used for.
Gecko = not so small (though still quite diddy), fast, very good at displaying HTML but also designed for groovy XML stuff like XUL - this adds its own overheads.
They are both pretty good, though Gecko is light years beyond KHTML in terms of technology. That's okay though, because KHTML does quite a good job of standard web pages and is fast to load/embed, unlike Gecko. KHTML for some reason also looks better on my box, I think it's a fonts issue.
And finally, if you want you can use the KMozilla bindings to replace KHTML with Gecko - well I say replace, actually unlike Windows KHTML is not required per se for KDE to function, rather an HTML Renderer with the correct KParts interface is. Somebody has wrapped those interfaces around Gecko and the result is a pluggable HTML handler. This is why browser integration with KDE is good (it's not actually specific to a browser/rendering engine) but with Windows it's bad (IE or else).
Wouldn't it make sense to try and shift some of the bandwidth pressure to the listeners? Look at it this way, broadband users normally have their upstream capacity idling. Why not have them retransmit the stream to a few modem users, or another broadband user for instance? Is there software that does this?
It's a cludgy hack though, what's really needed is for widespread deployment of multicast routers. That'd solve the problem nicely. Anybody know how likely this is to happen/timescales?
thanks -mike
* They'll have a longer product cycle (Macs last forever and retain their value much better than cobbled-together PCs), which means the district won't have to upgrade their hardware nearly as often.
LOL! So that's why Jobs just unveiled Quartz Extreme, an approach to speeding up the Aqua gui so extreme in itself that at the time of announcement, they were still selling laptops that could not run it! Apple regularly dump all over customers that are only a few years old.
# Macs are already entrenched in education for a good reason: they're ease of use is legendary. Quartz is, without a doubt, the best user interface ever.
Minor point - Quartz isn't a user interface, it's a graphics API. You mean Aqua, and personally I find it extremely unintuitive. "Ease of use" is largely subjective.
You know, we shouldn't just apply the principle of having only 1 product for each task to software! NO! We should apply it to everything.
Let's start with cars. Look at all those different makes and models: how much effort has been wasted by Ford, GM, Rover and the rest on designing cars that's - let's face it here guys - are all exactly the same.
That's the problem with the current state of car development - way too many companies, all in it for the money. Rather than having 1 all knowing monopoly making our cars, we have 10,000 different companies producing 10,000 models.
Maybe the only way to get car manufacturers organised is to nationalise the entire industry. Hell, why not do that for all industries? We can leave out software of course, that's already controlled by an all knowing monopoly.
What a radical idea. What should I call it? Oh wait.... that'd be communism...
I was thinking about this only 3 or 4 days ago.....
hehe. Ah well, I'm glad somebody else is doing it really, I have more than enough on my plate right now. Perhaps they should check out the Creative Commons?
You're missing the point. The reason most western governments aren't massively funding space exploration is because they are (by and large) run by a democracy that is forced to listen to the people - and right now, the attitude of the people is screw space, we have bigger problems, like 3rd world debt, growing the economy and reducing our dependance on oil.
As far as I'm concerned, it's a shame but exploring space is something my kids will have to do.
The desktop is much more challenging, but I don't actually see it as a "problem" if it becomes mainstream. It will certainly add pressure to improve standardisation work in the LSB for the user interface libraries.
That's a nice thought, but as there currently aren't that many distros that even manage to comply with the current LSB, how likely is it that UI libs will be standardised? I don't think it's possible to introduce standards into Linux now, it's too late for that. We just have to make do with whatever standards were set by the UNIX heritage, or by market forces.
I'm still trying to figure out if that's a good thing or not.
I agree, I don't think it's reasonable to ask the programmers to do anything any differently if they're already doing it for free. You do, however, always get what you pay for. Some software is free because people love the software, some is free because you couldn't GET anyone to buy it;-)
Hell, I've seen people try and charge for commercial software that looked like it was knocked together by a 15 year old in Visual Basic. Probably was in fact. Free software doesn't affect the quality at all, you can get high quality and low quality free software, just like commercial stuff.
I will, however, lay a good bit of blame on the "zealots" and the "community"
Please don't confuse the zealots with the community. Like any community made up on minorities, Linux has its share on zealots. They don't reflect the thinking of most Linux users or even developers. Criticism is allowed, even encouraged - Slashdot isn't where most development gets done. Go look at the KDE development lists for instance.
I wouldn't say it's spreading, far from it. It used to be the case that Linux could not be intelligently debated on Slashdot at all, now it can. Look at any recent topic on Linux, and you'll see plenty of measured, informed opinions. Nowadays the problem is with highly modded Mac zealots! There's no winning;)
About the extreme configurability thing: Linux was developed for power first, and ease of use second. This doesn't make it flawed in any way - Windows NT was power first, and only by the XP line was it easy to use, so it can be done.
Look at KDE - one of the new features in 3.1 will be (more) complete support for "kiosk" mode, which is basically controlling and removing features. It's intended for public/corporate computers, but could just as easily be used for home PCs too. Alternatively, distros can easily remove features and configurability.
I guess my point is, that it's better to have power and then control/hide it, than not to have it at all. That's why MS scrapped the 98 line and replaced it with XP - better to have power than not.
KDE, Gnome, and all the rest are chasing Microsoft because (deep breath here, folks)...Microsoft is where all the other folks WANT to be! They don't want to be buggy, huge, and expensive, but they do want to take advantage of the huge Windows penetration into the average Joe's computing experience. Linux folk consistently underestimate this factor, and then are puzzled why Linux is not gaining widespread acceptance on the desktop.
Ah, no, I don't agree with this bit. You're saying the reason desktop Linux hasn't taken off yet is because people think Windows is good enough. I disagree - people in my opinion don't think Windows is good enough. In fact, I often here even fairly techno-phobic people bitching about it: they've heard a techie swear at Microsoft when Windows crashed and they think: ah, I just lost all my work, this is the fault of Windows. And often they are right.
I'd say there are lots of good reasons why desktop Linux hasn't taken off yet. It's not because of any overarching problem with open source development, or any fundamental problem with Linux, it's just not there yet. People consistantly seem to underestimate how much work is required to make a truly easy to use computer, especially when the underlying OS was designed to be powerful first, simple second.
For instance: software management, fonts, printing/hardware setup, I could go on and on. There are currently several problems that mean that Linux is just too much hard work to use on the desktop right now. I can do it, and don't mind putting in the extra work because I like the "Free" aspect to it. But most don't care. Combine this with small mindshare and the difficulty in getting preinstalls, and I think it's self evident why Linux hasn't got there yet. But it will.
Cardinal rule: a product does NOT have to be the BEST at anything, it just has to be GOOD ENOUGH, and CONVENIENT ENOUGH, to get the job done MOST OF THE TIME......... Intel's processors are not the fastest, they do not have the most elegant design, and they sure as hell aren't the cheapest, but they rule the PC world completely while Alpha and PowerPC occupy niches. Macintosh arguably has one of the better GUI's out there, yet they languish with only about 4% of the market. OS/2 was 32-bits long before WindowsNT was even a glimmer in the eye, but does anybody really run Warp anymore?
Intel - this was mainly the case because Intel were in the right place at the right time, and because they did the famous Wintel deal.
Macintosh - yeah, for years they had the best GUIs, but let's face it, up until recently Macs sucked at everything else. OS 9 made Windows 98 look like a magical technological feat of engineering. There was a reason their market share dropped so rapidly.
OS/2 - the reasons for the lack of dominance here are well documented, and they aren't to do with technology.
My point is, none of these things are gone today because Windows did what they did better: there were almost other (usually business related) reasons involved.
Being the BEST at one thing frequently means you've neglected something else somewhere. Linux is a technological marvel in its configurability and flexibility, but has neglected usability with respect to contemporary products from Microsoft and Apple. Don't try to deny it, it's true.
It is true. However, that's because the Linux developers first concentrated on power, then flexibility, then features, and are just now turning their focus onto usability and looks. For instance, in the last 2 weeks the KDE Panel configuration dialog was patched to make it far more usable, with a cleaner, simpler design. Now Linux on the desktop has caught up with Windows in terms of raw features, it's busy taming them and making them easier to use. Meanwhile, MS and Apple worked on ease of use above all else, and are now focussing on power and features. Take for example the pushes they are making into the server arenas etc. Linux and the commerical OSes have just approached things from a different angle.
When your grandmother can successfully get a PC and load RedHat on it unassisted, and then actually troubleshoot it if something goes wrong (can she understand cronjobs? fsck?)
She doesn't have to? If anything right now Linux is more granny friendly than Windows (98) because they rarely set things up themselves, rather they just read email, write letters etc. With Linux once it's setup it's hard to screw up (you need root access). And more to the point, I don't need to know about fsck or cronjobs, they are dealt with automatically by SuSE. Linux still isn't user friendly enough for widespread adoption, but it's getting closer.
Unfortunately, I have a funny feeling that in order to become that user friendly, Linux will have to become bigger, slower, more expensive, and more proprietary. Perhaps it isn't true, but I'd be willing to bet that it is.
I'd bet against that. So far Linux has come on in leaps and bounds, and it's remained cheap, free and... well okay I give it to you on speed grounds.
Hmm, if I were you I'd try Linux again perhaps in a few months, but this time try putting some more effort in. I spent 10 minutes on kde-look.org and looroll.com and came away with a beautiful and unique looking desktop. The default KDE look -is- ugly, for KDE3.1 I think they may be planning to move away from that look towards some of the more advanced ones coming online now, such as Crystal/Keramik.
By "Fonts not displaying properly" I'd guess you mean they look awful. This is true. The fonts situation is improving rapidly however, as programs switch to the new font systems in XFree 4.2 and FreeType.
Websites not working properly: I'm afraid if you absolutely MUST use Windows Media (why? aren't you at work?) then yes you are a bit stuffed. Remember though that the "free" tools you get with Windows or the Mac aren't actually free, they are included in the price of Windows. Try looking at it that way.
Finally, remember that if you stick with Windows now, you'll be stuffed in future - try looking at the license agreements. If you go with Linux, yes, there will be some annoying problems, but they are being resolved fast, and more importantly you'll be in a better position in the future. You'll also have a competitive advantage over other companies still paying for MS wares.
They are two guys who have written a complete Outlook clone (well sort of) in DHTML. It works really well (if you're an IE only sort of person) and costs about 30 quid a year iirc. You also get POP and IMAP access. Their DHTML webmail client blows Hotmail away, it's got stuff like drag and drop, just like a software email client.
my mac.com email is free, web-based, devoid of banners, and apple probably wouldn't pull this kinda crap.
I remember a friend of mine tried to get me to use a mac.com account: it wouldn't let me use it though, because I didn't have a Mac. It didn't surprise me, the mac.com services are basically included in the price of the Apple hardware.
A week ago said friend told me they no longer locked out people who didn't use Macs - but as far as I'm concerned they already told me to get lost in the least subtle way possible, so why should I bother with them?
The browser wars only led to feature bloat because two large corporations were cat-fighting. These days there is a much greater emphasis on standards: even IE is getting in on the act, slowly.
Yes, competition is good, but there is a point where there are just too many browsers. As a maker of all things web, it is very difficult to work around all the quirks of these browsers (and yes, *all* browsers have quirks)
That's like saying there are too many makes of car. Nobody says that, because all cars conform to some basic defacto standards in terms of height, roadwidth etc.
The solution is to have much better standards compliance. Mozilla is spot on. Opera I think lacked DOM support, not sure if that's fixed, I'd guess it is by now. KHTML is also pretty good. IE sucks of course, but IE6 is getting there (level 1 while moz crew start on level 3)
Slashdot Mirror
Most proprietary file formats are by nature binary though, how would you represent say an image as XML? You can't of course, and often files are mixed textual/binary such as MS Word files.
Really I think there is a line to be drawn here - sometimes (as in the case with MS) file formats are being kept secret to facilitate lockin. This is clearly unacceptable, as it impedes the economy and restricts competition. I seriously doubt the MS Word file format contains vast amounts of innovation that everyone could steal if they were forced to document it.
On the other hand, especially with codecs for instance, there COULD be a lot of intellectual property involved that has cost time and money to research. Should they be forced to open up as well? Clearly not. Really this discussion is about, where to find the balance.
Really I think before machine translation is a workable reality, we'll need something like Cyc. To get a real, decent translation simple rule based systems aren't sufficient - you need to be able to translate the languages into logical formulae that can then be reconstructed into the new language. Cyc already has this ability to a limited extent (though only for english), but really, I think this is the way to go.
Let's face it, who could resist the idea of getting a cool computer while at the same time losing Microsoft money? It's a fab idea!
You're right, but unfortunately techniques for writing cross platform games are currently immature. SDL is about as good as it gets, and it isn't yet up to the standard of DirectX. Also, until OpenGL gets significantly ahead of Direct3D, games companies will continue to write to D3D, and therefore card manufacturers will write better Direct3D drivers (or no opengl drivers) as that's where the money is.
Steps to helping developers write cross platform games:
Read the whitepaper - GNOME 2 support is in there, and in the PDF is says actually only KDE Base is mandantory. Some of the details seem a little inconsistent right now, I think they're still figuring out exactly what it'll be.
Actually according to the white paper, there will be RedHat compatible libs/symlinks ,so in theory 1 RPM will work on 5 distros. I'm not sure I like this however, as it'd effectively make RedHat the "standard" to which RPMs would be targeted, as then you know it'll work natively on redhat, and with the compatability mode on UL systems.
Hmm, I'm pretty sure ALSA is included in most distros these days. SuSE 8 I know has ALSA 0.9 in it, which is pretty good. I'm using ALSA now in fact and I'm on 7.3 Just because the official kernel doesn't have something doesn't mean it won't be there remember.
RedHat derived distros like Mandrake are also seeing problems with stuff like file paths and so on.
Car analogies are often flawed - remember that cars already ARE standardised in that they all are roughly the same width, height etc. They all have similar turning circles: basically they all work on our roads in the same way. In computers we have a situation where some cars can only go down some roads and it's a mess. Roll on LSB.
This is one that I noticed on the KMail homepage. If you use KMail with IMAP then please please upgrade for your own sake, or else massive dataloss could occur! I'm not sure how the KMail team let this one slip past, but it's a pretty important fix.
KHTML = small, light, fast, good at displaying HTML quickly. Ideal for embedding, which is what it's used for.
Gecko = not so small (though still quite diddy), fast, very good at displaying HTML but also designed for groovy XML stuff like XUL - this adds its own overheads.
They are both pretty good, though Gecko is light years beyond KHTML in terms of technology. That's okay though, because KHTML does quite a good job of standard web pages and is fast to load/embed, unlike Gecko. KHTML for some reason also looks better on my box, I think it's a fonts issue.
And finally, if you want you can use the KMozilla bindings to replace KHTML with Gecko - well I say replace, actually unlike Windows KHTML is not required per se for KDE to function, rather an HTML Renderer with the correct KParts interface is. Somebody has wrapped those interfaces around Gecko and the result is a pluggable HTML handler. This is why browser integration with KDE is good (it's not actually specific to a browser/rendering engine) but with Windows it's bad (IE or else).
It's a cludgy hack though, what's really needed is for widespread deployment of multicast routers. That'd solve the problem nicely. Anybody know how likely this is to happen/timescales? thanks -mike
LOL! So that's why Jobs just unveiled Quartz Extreme, an approach to speeding up the Aqua gui so extreme in itself that at the time of announcement, they were still selling laptops that could not run it! Apple regularly dump all over customers that are only a few years old.
# Macs are already entrenched in education for a good reason: they're ease of use is legendary. Quartz is, without a doubt, the best user interface ever.
Minor point - Quartz isn't a user interface, it's a graphics API. You mean Aqua, and personally I find it extremely unintuitive. "Ease of use" is largely subjective.
You know, we shouldn't just apply the principle of having only 1 product for each task to software! NO! We should apply it to everything.
Let's start with cars. Look at all those different makes and models: how much effort has been wasted by Ford, GM, Rover and the rest on designing cars that's - let's face it here guys - are all exactly the same.
That's the problem with the current state of car development - way too many companies, all in it for the money. Rather than having 1 all knowing monopoly making our cars, we have 10,000 different companies producing 10,000 models.
Maybe the only way to get car manufacturers organised is to nationalise the entire industry. Hell, why not do that for all industries? We can leave out software of course, that's already controlled by an all knowing monopoly.
What a radical idea. What should I call it? Oh wait.... that'd be communism...
I was thinking about this only 3 or 4 days ago.....
hehe. Ah well, I'm glad somebody else is doing it really, I have more than enough on my plate right now. Perhaps they should check out the Creative Commons?
Who said anything about foreign oil? I was talking about -all- oil, and wasn't especially thinking about America at all, in fact I'm a European.
As far as I'm concerned, it's a shame but exploring space is something my kids will have to do.
That's a nice thought, but as there currently aren't that many distros that even manage to comply with the current LSB, how likely is it that UI libs will be standardised? I don't think it's possible to introduce standards into Linux now, it's too late for that. We just have to make do with whatever standards were set by the UNIX heritage, or by market forces.
I'm still trying to figure out if that's a good thing or not.
Hell, I've seen people try and charge for commercial software that looked like it was knocked together by a 15 year old in Visual Basic. Probably was in fact. Free software doesn't affect the quality at all, you can get high quality and low quality free software, just like commercial stuff.
I will, however, lay a good bit of blame on the "zealots" and the "community"
Please don't confuse the zealots with the community. Like any community made up on minorities, Linux has its share on zealots. They don't reflect the thinking of most Linux users or even developers. Criticism is allowed, even encouraged - Slashdot isn't where most development gets done. Go look at the KDE development lists for instance.
I wouldn't say it's spreading, far from it. It used to be the case that Linux could not be intelligently debated on Slashdot at all, now it can. Look at any recent topic on Linux, and you'll see plenty of measured, informed opinions. Nowadays the problem is with highly modded Mac zealots! There's no winning ;)
About the extreme configurability thing: Linux was developed for power first, and ease of use second. This doesn't make it flawed in any way - Windows NT was power first, and only by the XP line was it easy to use, so it can be done.
Look at KDE - one of the new features in 3.1 will be (more) complete support for "kiosk" mode, which is basically controlling and removing features. It's intended for public/corporate computers, but could just as easily be used for home PCs too. Alternatively, distros can easily remove features and configurability.
I guess my point is, that it's better to have power and then control/hide it, than not to have it at all. That's why MS scrapped the 98 line and replaced it with XP - better to have power than not.
Ah, no, I don't agree with this bit. You're saying the reason desktop Linux hasn't taken off yet is because people think Windows is good enough. I disagree - people in my opinion don't think Windows is good enough. In fact, I often here even fairly techno-phobic people bitching about it: they've heard a techie swear at Microsoft when Windows crashed and they think: ah, I just lost all my work, this is the fault of Windows. And often they are right.
I'd say there are lots of good reasons why desktop Linux hasn't taken off yet. It's not because of any overarching problem with open source development, or any fundamental problem with Linux, it's just not there yet. People consistantly seem to underestimate how much work is required to make a truly easy to use computer, especially when the underlying OS was designed to be powerful first, simple second.
For instance: software management, fonts, printing/hardware setup, I could go on and on. There are currently several problems that mean that Linux is just too much hard work to use on the desktop right now. I can do it, and don't mind putting in the extra work because I like the "Free" aspect to it. But most don't care. Combine this with small mindshare and the difficulty in getting preinstalls, and I think it's self evident why Linux hasn't got there yet. But it will.
Cardinal rule: a product does NOT have to be the BEST at anything, it just has to be GOOD ENOUGH, and CONVENIENT ENOUGH, to get the job done MOST OF THE TIME. ........ Intel's processors are not the fastest, they do not have the most elegant design, and they sure as hell aren't the cheapest, but they rule the PC world completely while Alpha and PowerPC occupy niches. Macintosh arguably has one of the better GUI's out there, yet they languish with only about 4% of the market. OS/2 was 32-bits long before WindowsNT was even a glimmer in the eye, but does anybody really run Warp anymore?
Intel - this was mainly the case because Intel were in the right place at the right time, and because they did the famous Wintel deal.
Macintosh - yeah, for years they had the best GUIs, but let's face it, up until recently Macs sucked at everything else. OS 9 made Windows 98 look like a magical technological feat of engineering. There was a reason their market share dropped so rapidly.
OS/2 - the reasons for the lack of dominance here are well documented, and they aren't to do with technology.
My point is, none of these things are gone today because Windows did what they did better: there were almost other (usually business related) reasons involved.
Being the BEST at one thing frequently means you've neglected something else somewhere. Linux is a technological marvel in its configurability and flexibility, but has neglected usability with respect to contemporary products from Microsoft and Apple. Don't try to deny it, it's true.
It is true. However, that's because the Linux developers first concentrated on power, then flexibility, then features, and are just now turning their focus onto usability and looks. For instance, in the last 2 weeks the KDE Panel configuration dialog was patched to make it far more usable, with a cleaner, simpler design. Now Linux on the desktop has caught up with Windows in terms of raw features, it's busy taming them and making them easier to use. Meanwhile, MS and Apple worked on ease of use above all else, and are now focussing on power and features. Take for example the pushes they are making into the server arenas etc. Linux and the commerical OSes have just approached things from a different angle.
When your grandmother can successfully get a PC and load RedHat on it unassisted, and then actually troubleshoot it if something goes wrong (can she understand cronjobs? fsck?)
She doesn't have to? If anything right now Linux is more granny friendly than Windows (98) because they rarely set things up themselves, rather they just read email, write letters etc. With Linux once it's setup it's hard to screw up (you need root access). And more to the point, I don't need to know about fsck or cronjobs, they are dealt with automatically by SuSE. Linux still isn't user friendly enough for widespread adoption, but it's getting closer.
Unfortunately, I have a funny feeling that in order to become that user friendly, Linux will have to become bigger, slower, more expensive, and more proprietary. Perhaps it isn't true, but I'd be willing to bet that it is.
I'd bet against that. So far Linux has come on in leaps and bounds, and it's remained cheap, free and ... well okay I give it to you on speed grounds.
By "Fonts not displaying properly" I'd guess you mean they look awful. This is true. The fonts situation is improving rapidly however, as programs switch to the new font systems in XFree 4.2 and FreeType.
Websites not working properly: I'm afraid if you absolutely MUST use Windows Media (why? aren't you at work?) then yes you are a bit stuffed. Remember though that the "free" tools you get with Windows or the Mac aren't actually free, they are included in the price of Windows. Try looking at it that way.
Finally, remember that if you stick with Windows now, you'll be stuffed in future - try looking at the license agreements. If you go with Linux, yes, there will be some annoying problems, but they are being resolved fast, and more importantly you'll be in a better position in the future. You'll also have a competitive advantage over other companies still paying for MS wares.
They are two guys who have written a complete Outlook clone (well sort of) in DHTML. It works really well (if you're an IE only sort of person) and costs about 30 quid a year iirc. You also get POP and IMAP access. Their DHTML webmail client blows Hotmail away, it's got stuff like drag and drop, just like a software email client.
I remember a friend of mine tried to get me to use a mac.com account: it wouldn't let me use it though, because I didn't have a Mac. It didn't surprise me, the mac.com services are basically included in the price of the Apple hardware.
A week ago said friend told me they no longer locked out people who didn't use Macs - but as far as I'm concerned they already told me to get lost in the least subtle way possible, so why should I bother with them?
The browser wars only led to feature bloat because two large corporations were cat-fighting. These days there is a much greater emphasis on standards: even IE is getting in on the act, slowly.
That's like saying there are too many makes of car. Nobody says that, because all cars conform to some basic defacto standards in terms of height, roadwidth etc.
The solution is to have much better standards compliance. Mozilla is spot on. Opera I think lacked DOM support, not sure if that's fixed, I'd guess it is by now. KHTML is also pretty good. IE sucks of course, but IE6 is getting there (level 1 while moz crew start on level 3)