Pascal was designed as a teaching language, its small and tight and easy to understand. It doesn't have the wealth of libraries that other languages have, and perhaps that's a point in its favour.
Still, Delphi was incredibly popular until recently, perhaps being used as a smartphone language would revive it, and a love of programming too.
Python has too many quirks to make it good for true beginners. You have to make entry incredibly easy for them to keep with it - which is why PHP is so popular.
amen. Pascal was designed as a teaching language. The only reason it didn't catch on for business apps was because it didn't have a standard library. Something that was rectified with Turbo Pascal and Delphi. And look how popular they became
Perhaps the BASIC for smartphones should be a Pascal renaissance, Whoever owns Delphi might like to think about that and get Google to preinstall it on all Androids:)
Office scripting is done with VB.net because of the greater COM integration with that language. C# and VB might be the same thing under the covers but they focus on different tasks. Nobody sensible would write Office scripts in C# when there's a system that's better designed for that type of task.
that's true. Though, I don't think its the language so much that's any easier (except that whitespace thing in Python that's going to annoy a beginner a lot - you need to make it easy, and not realising your code's broken because you're forced to be disciplined about whitespace is not going to help you focus on the difficult, new coding problem).
The problem is the ecosystem of Python. With PHP you install (or its already there usually) and off you go. My experience with python is that it requires one of 3 different versions of the language, eggs that are difficult to obtain and/or compile and an environment that doesn't easily configure in Apache.
That's the main reason PHP is really successful. When you're an 'amateur' programmer, any obstacles to getting your code working are much larger than we realise.
not quite. Capitalism would have destroyed the entire economic system in the last few years. Fortunately, the governments and their taxpayer funds were there to bail out the banks and prop the whole thing up.
Capitalism is just as flawed as the rest of the systems. To point out one above the others is an ignorant choice. However, the others are flawed too.
The world is a complex place, to try and choose one system to rule is too simplistic to really work. What you need is to take from all of them, mix it up a bit and try to tweak it as you go to ensure smooth working of the system.
So unfettered capitalism would have the peasant workers killed for being unproductive while a few super-rich got richer and richer and more and more inbred and stupid. You'd end up with a truly stagnant society where the talented poor could do nothing to fix the problems and the decadent rich cannot conceive of change.
So you have capitalism for the majority, but then ensure its regulated so the poor do not starve (socialism!), the rich do not take over the world (communism!) and everyone has to abide by rules that are enforced by others (facism!)
After all, you say "capitalism has allowed the US to provide more humanitarian assistance"... a) that has no place in a capitalist system - charity is a socialist concept (unless you're subsidising a market so they can purchase more of your stuff, but that hardly applies to dictatorial African states.. unless they're buying arms of course... hmm), b) the US is technically poorer that most places, the only reason you're not the recipient of humanitarian aid is because you've received it in the form of debt (so it doesn't count as aid). if you had any chance of paying off the national debt, then maybe you'd have a point. Chances are, China is going to be paying your way for you for some time.
I doubt anyone's made any study of this, but children's books have been around for a long time and adults have gone on from them to more 'serious' work.
for example, Dr Seuss's works have been about for a long time, lots of children have read them, yet you're not seriously arguing that adults of today cannot read anything that doesn't rhyme in simple stanzas.
There has to be a complex ecosystem of books, suitable for everyone's level of reading ability. That means everything from the Gruffalo to Harry Potter to Philip Pullman to CS Lewis to Isaac Asimov to Tolkien. Some kids will start with at the bottom and never progress further - but they probably wouldn't have progressed anyway, what with today's education system not pushing them sufficiently. But there are others who will read on, and though you could say they would have done so without the Harry Potters, they certainly couldn't if the books they want to read didn't exist. Hence, you need the Potters (and all the others) to be there for them.
Don't be too snobbish about the quality of literary works. I'd say the readability of CS Lewis and Potter to be about the same. Sure the concepts contained in there are different, but many 10 year olds wouldn't grasp them anyway.
not really. we have plenty of paper and press time to print books, shipping is far more plentiful than you realise.
For ebooks, the analogies are bandwidth (=shipping), and cpu time (=press capacity). So while you may think that ebooks are totally free of any constraints, they're not. Similarly, paper books are not so expensive that it makes any practical difference.
The only thing that does matter here is the cost of paper and the storage requirements. These only affect the quantity of books, so a virtual library will contain everything whereas a physical library will only contain a subset of all work. (but that's not so bad as you end up with an inter-library lending system).
Apart from the capitalist system of payment for popularity, other works can be paid for in alternative means - eg universities funding research which are published in book form. Vested interests publishing their 'propaganda'. In both cases, the success of the end book is meaningless. That it exists for others to read is the end-goal, as the author is paid for the initial work, not the success of the book.
and the other side of that is that new stuff isn't always better by a long way.
I mean, look at the tools we're using to connect to this site - still using ethernet? surely we should have scrapped that ancient technology by now.... and the move towards thin clients with all the data held on the 'cloud'. Isn't that just mainframe style development all over again?
A lot of the old guys will tell you that something is better, not because they're "stuck in the past" but because the techniques they're talking about really are better. There are too many 'latest fads' in IT today, often they become the biggest hyped up thing ever, and after a year or two everyone recognises that they were just bull.
Ok, sure there are old guys who do reminisce about the past too much, but by the same token there are too many young guys who think that everything the currently exists is rubbish because they can do it better.
The industry really needs to grow up and understand that building on what has gone before is beneficial, not to (continually) scrap it and start over again.
but are they, if the "cheap talent" walks out the door taking all that knowledge with them, leaving you having to hire someone new, then train them up. It can take a long time to get someone up to speed with a product, and even longer to get them to really know its ins and outs. I'd say a year easily for most of the complex apps that are undocumented and chaotically developed by the last guys who walked away from it.
The TCO of staff should not be underestimated by the management, they use such bullsh*t to explain why Windows is better than Linux, but they won't apply the same to staff.
yeah, you'd have thought that devs would have "grown up" by now, focussing more on end-product with all the responsibilities that entails rather than a somewhat teenage approach that is more based on the developer's egos.
Maybe in another ten years, but I imagine the current crop of devs who participate in this environment will give up, leaving it to a new bunch of kids who perpetuate the same old attitudes.
why do all that work (loading, configuring, and then swapping it to disk) just so you have something that you don't use.
Its sloppy, lazy, concepts like this that may be easier for the developer, but make startup of the OS slow. Ever wondered what all all that disk activity was when starting your OS was? It was reading stuff from disk, then writing it straight back. Like the old make-work schemes that got people digging holes then filling them back in again.
The other reason is that once you develop your system like this, once you want to print, you'll end up re-loading all that print subsystem into memory, even if you decide not to print anything. Ok, that's a poor example, but you do get other apps that do lots of swapping all the time to get tiny bits of functionality loaded.
you do realise the goatse man was a product of this natural selection process too.... I don't think you can hold us up as an example of perfection quite yet (not until we've achieved transcendance like today's motd says)
I remember when Microsoft put the close button on the windows, there was a lot of wailing and gnashing of teeth as people claimed they'd always accidentally close the window instead of maximising it... and I'm sure there were, but people quickly got used to it.
Change it to something else, there will be much wailing again, but they'll get used to it readily enough again.
(I quite like the idea of moving the title bar to the side, but on the right hand side, as a 'handle' like the ones you get on all appliances that are designed for right-handed people).
amen to that. Perhaps it is the proliferation of programming styles that adds so much weight to a system like this, and if everything was coded in a single script language and a single compiled language the whole would be a lot lighter and faster.
That, and a good separation of concerns for all modules - the printing subsystem shouldn't even be loaded until you need to print something, which means the design needs to understand what you can send to the print subsystem without being built with it included (ie a set of protocol standards)
I'm amazed that they beat the financial crisis. We all know Windows costs a lot for licences and today's governments are very cash-strapped. So like Portugal that recently announced they would not pay for any more Windows upgrades, I'm surprised more countries aren't looking closely at Munich to see if they can reduce their deficits slightly by going this route too.
imagine the history function is a method that's called. Does it really need PGO to speed up a function call? Or even if the display routines store up all links in a page then send them to a history function which returns all the ones that have been visited to be re-coloured after the final display is calculated, really does that need PGO? If one takes.25 seconds to run, would any call optimisation really shave any meaningful time off that number? The history module would still be optimised within itself after all, and just because the code is in a module doesn't mean it would have to manage the history data totally separately at runtime.
That's what I'm talking about here, and it could well be that the rendering engine is so huge it screws the optimiser by itself, it doesn't mean that a large majority of the non-critical-path codebase could be modularised, ie all the stuff surrounding the page rendering.
Of course, if such a small modification does make a large performance hit, then you have to wonder what the code paths are like. A certain amount of simplifying might make a much larger difference in such a case.
either C/C++ where you build for each platform or.NET where you only have 1 platform, but the user still has to download the required.net runtime and security patches. or Java where the user has to download the required JRE
You're probably best off with the Linux model where the code is downloaded and compiled on whatever platform the user is running.
I think the error message thing is just for templates too, as they're notorious to make easy to read. I've seen error messages for java apps where you have a stack trace that tells you nothing useful, and.NET exceptions that boil down to a Win32 structured exception code. None can be described as perfect.
So, yes, quite right - no language, including C++, is perfect. The FQA doesn't take that view though, it assumes C++ is the worst thing ever and doesn't try to hide its lack of objectivity. He hasn't exactly done a FQA for VB6 or PHP now has he.
I suppose you could use a VM based language where the optimisation takes place at run time instead, but they're trying to do all that in one step where it affects the developer (who can take the need to have a super computer to build) rather than the user (who doesn't want to have a super computer to run it), or an interpreted language where you have to write a VM to run it to get any form of acceptable performance.
Or you write it in something like C++ and complain that the optimiser tool uses too much memory itself (or that the developers didn't modularise it sufficiently)
Either way, the FQA is a terrible rant with plenty of over-exaggerations, inconsistencies and 'bad things' that were obsoleted decades ago. Ignore it and hope it goes away, or read this for some rebuttals.
I don't suppose the huge web-display system needs to be split into pieces (though I imagine there's not that much overhead to it - you won't get any PGO, but that's not that big an optimisation step as you'll still get it inside each module).
What could be done is to split the rest of the browser into modules. The code that displays a window doesn't have to be in the same module as the code that figures out how to display the data stream. The bookmarks system could be a separate module that's linked in at the end - its hardly essential that something like that is optimised to the operation of the whole.
disagree (and I'm not an Apple fanboi BTW). I think people here hate Apple becuase of their closed-in attitude. You buy an iPod and you only (effectively) can access it with approved Apple stuff. You buy a iPhone and you're locked into their store, etc etc etc.
We have to give Apple credit for kickstarting the whole smartphone industry and changing the world. You have to give them credit for popularising GUI interfaces and similar.
You also have to criticise them for the lock in and overpricing though, but their products aren't garbage (or no-one would keep buying them).
Naturally, bringing up Office means any UI arguments are automatically invalidated:) Office did things differently to everyone else for some reason, and everyone accepted and grumbled about it. But all the other apps were of a consistent standard. This was cited as part of Windows' lower TCO, and there was a book on how to make your application conform to the Windows UI style guidelines.
I think today's UI styles are back in the cowboy era of web design where you can do anything anywhere you like, and so designers do exactly that to differentiate themselves. I also think that they'll start to conform again when people have got used to the new stuff, or get so annoyed with geocities-style interfaces.
Re:Has he ever actually talked to users?
on
The Condescending UI
·
· Score: 5, Insightful
which is exactly why the new UIs are so poor.
Remember when Windows first came out, it has menus and every app had the same menu bar. Everything had a file menu that had new/open/print/etc items on it.
You could open any app and instantly know how to create a new document - because there was the file|new menu item, every time. You'd received training for all apps, instant familiarity, instant productivity.
Fast forward to today and we have different interfaces for everything. The new UIs with shiny orbs and animated transitions mean you have to figure out where all the new bits are for each app. Then some of them start working differently (eg Excel that has multiple icons in the task bar, but they're all running in a single instance so you close 1 you close them all kind of bo**ocks), and some don't even have menus - well, they have menus, but they're tucked away behind a little coloured icon so they appear when you click it, if you can find the f***er in the first place (eg the new hide-everything-away browser interfaces). The the ribbon comes along (which is a fine toolbar repacement BTW) but is used as a menu replacement too - with loads of bits hidden away in little menus behind tiny ">" icons.
The old interfaces were fugly, but functional. They made us productive and really that's what is needed for line-of-business apps. No-one really cares that excel looks cool, not when you're typing in the accounts.
sure, but MS supports all end-of-life products for 10 years, even VB6 was supported until recently.
What you get for your support however, generally means security bugs are squashed and MS will spend a tiny amount of effort making your apps work on future versions of Windows, if they feel like it. If you have a problem they'll tell you to install the latest service pack and if that doesn't fix it, tough - it doesn't support what you want it to do.
yep, Epson has one too, though it won't print pdfs for some reason. I guess because they expect you to just use it to send images to your printer. However, I use a WP4545 wireless printer and you just send whatever you like to it, from anywhere.
Epson also comes with a email printing service - you send your document to your own, unique email address and it forwards it to your printer. I don't have it set up, but it seems a nice idea.
Slashdot Mirror
Pascal was designed as a teaching language, its small and tight and easy to understand. It doesn't have the wealth of libraries that other languages have, and perhaps that's a point in its favour.
Still, Delphi was incredibly popular until recently, perhaps being used as a smartphone language would revive it, and a love of programming too.
Python has too many quirks to make it good for true beginners. You have to make entry incredibly easy for them to keep with it - which is why PHP is so popular.
amen. Pascal was designed as a teaching language. The only reason it didn't catch on for business apps was because it didn't have a standard library. Something that was rectified with Turbo Pascal and Delphi. And look how popular they became
Perhaps the BASIC for smartphones should be a Pascal renaissance, Whoever owns Delphi might like to think about that and get Google to preinstall it on all Androids :)
Office scripting is done with VB.net because of the greater COM integration with that language. C# and VB might be the same thing under the covers but they focus on different tasks. Nobody sensible would write Office scripts in C# when there's a system that's better designed for that type of task.
that's true. Though, I don't think its the language so much that's any easier (except that whitespace thing in Python that's going to annoy a beginner a lot - you need to make it easy, and not realising your code's broken because you're forced to be disciplined about whitespace is not going to help you focus on the difficult, new coding problem).
The problem is the ecosystem of Python. With PHP you install (or its already there usually) and off you go. My experience with python is that it requires one of 3 different versions of the language, eggs that are difficult to obtain and/or compile and an environment that doesn't easily configure in Apache.
That's the main reason PHP is really successful. When you're an 'amateur' programmer, any obstacles to getting your code working are much larger than we realise.
not quite. Capitalism would have destroyed the entire economic system in the last few years. Fortunately, the governments and their taxpayer funds were there to bail out the banks and prop the whole thing up.
Capitalism is just as flawed as the rest of the systems. To point out one above the others is an ignorant choice. However, the others are flawed too.
The world is a complex place, to try and choose one system to rule is too simplistic to really work. What you need is to take from all of them, mix it up a bit and try to tweak it as you go to ensure smooth working of the system.
So unfettered capitalism would have the peasant workers killed for being unproductive while a few super-rich got richer and richer and more and more inbred and stupid. You'd end up with a truly stagnant society where the talented poor could do nothing to fix the problems and the decadent rich cannot conceive of change.
So you have capitalism for the majority, but then ensure its regulated so the poor do not starve (socialism!), the rich do not take over the world (communism!) and everyone has to abide by rules that are enforced by others (facism!)
After all, you say "capitalism has allowed the US to provide more humanitarian assistance"... a) that has no place in a capitalist system - charity is a socialist concept (unless you're subsidising a market so they can purchase more of your stuff, but that hardly applies to dictatorial African states.. unless they're buying arms of course... hmm), b) the US is technically poorer that most places, the only reason you're not the recipient of humanitarian aid is because you've received it in the form of debt (so it doesn't count as aid). if you had any chance of paying off the national debt, then maybe you'd have a point. Chances are, China is going to be paying your way for you for some time.
I doubt anyone's made any study of this, but children's books have been around for a long time and adults have gone on from them to more 'serious' work.
for example, Dr Seuss's works have been about for a long time, lots of children have read them, yet you're not seriously arguing that adults of today cannot read anything that doesn't rhyme in simple stanzas.
There has to be a complex ecosystem of books, suitable for everyone's level of reading ability. That means everything from the Gruffalo to Harry Potter to Philip Pullman to CS Lewis to Isaac Asimov to Tolkien. Some kids will start with at the bottom and never progress further - but they probably wouldn't have progressed anyway, what with today's education system not pushing them sufficiently. But there are others who will read on, and though you could say they would have done so without the Harry Potters, they certainly couldn't if the books they want to read didn't exist. Hence, you need the Potters (and all the others) to be there for them.
Don't be too snobbish about the quality of literary works. I'd say the readability of CS Lewis and Potter to be about the same. Sure the concepts contained in there are different, but many 10 year olds wouldn't grasp them anyway.
not really. we have plenty of paper and press time to print books, shipping is far more plentiful than you realise.
For ebooks, the analogies are bandwidth (=shipping), and cpu time (=press capacity). So while you may think that ebooks are totally free of any constraints, they're not. Similarly, paper books are not so expensive that it makes any practical difference.
The only thing that does matter here is the cost of paper and the storage requirements. These only affect the quantity of books, so a virtual library will contain everything whereas a physical library will only contain a subset of all work. (but that's not so bad as you end up with an inter-library lending system).
Apart from the capitalist system of payment for popularity, other works can be paid for in alternative means - eg universities funding research which are published in book form. Vested interests publishing their 'propaganda'. In both cases, the success of the end book is meaningless. That it exists for others to read is the end-goal, as the author is paid for the initial work, not the success of the book.
I was trying to be generous :)
and avoid turning it into an 'lets outsource to India, they have programmers too and they're cheaper" discussion.
and the other side of that is that new stuff isn't always better by a long way.
I mean, look at the tools we're using to connect to this site - still using ethernet? surely we should have scrapped that ancient technology by now.... and the move towards thin clients with all the data held on the 'cloud'. Isn't that just mainframe style development all over again?
A lot of the old guys will tell you that something is better, not because they're "stuck in the past" but because the techniques they're talking about really are better. There are too many 'latest fads' in IT today, often they become the biggest hyped up thing ever, and after a year or two everyone recognises that they were just bull.
Ok, sure there are old guys who do reminisce about the past too much, but by the same token there are too many young guys who think that everything the currently exists is rubbish because they can do it better.
The industry really needs to grow up and understand that building on what has gone before is beneficial, not to (continually) scrap it and start over again.
but are they, if the "cheap talent" walks out the door taking all that knowledge with them, leaving you having to hire someone new, then train them up. It can take a long time to get someone up to speed with a product, and even longer to get them to really know its ins and outs. I'd say a year easily for most of the complex apps that are undocumented and chaotically developed by the last guys who walked away from it.
The TCO of staff should not be underestimated by the management, they use such bullsh*t to explain why Windows is better than Linux, but they won't apply the same to staff.
yeah, you'd have thought that devs would have "grown up" by now, focussing more on end-product with all the responsibilities that entails rather than a somewhat teenage approach that is more based on the developer's egos.
Maybe in another ten years, but I imagine the current crop of devs who participate in this environment will give up, leaving it to a new bunch of kids who perpetuate the same old attitudes.
why do all that work (loading, configuring, and then swapping it to disk) just so you have something that you don't use.
Its sloppy, lazy, concepts like this that may be easier for the developer, but make startup of the OS slow. Ever wondered what all all that disk activity was when starting your OS was? It was reading stuff from disk, then writing it straight back. Like the old make-work schemes that got people digging holes then filling them back in again.
The other reason is that once you develop your system like this, once you want to print, you'll end up re-loading all that print subsystem into memory, even if you decide not to print anything. Ok, that's a poor example, but you do get other apps that do lots of swapping all the time to get tiny bits of functionality loaded.
you do realise the goatse man was a product of this natural selection process too.... I don't think you can hold us up as an example of perfection quite yet (not until we've achieved transcendance like today's motd says)
I remember when Microsoft put the close button on the windows, there was a lot of wailing and gnashing of teeth as people claimed they'd always accidentally close the window instead of maximising it... and I'm sure there were, but people quickly got used to it.
Change it to something else, there will be much wailing again, but they'll get used to it readily enough again.
(I quite like the idea of moving the title bar to the side, but on the right hand side, as a 'handle' like the ones you get on all appliances that are designed for right-handed people).
amen to that. Perhaps it is the proliferation of programming styles that adds so much weight to a system like this, and if everything was coded in a single script language and a single compiled language the whole would be a lot lighter and faster.
That, and a good separation of concerns for all modules - the printing subsystem shouldn't even be loaded until you need to print something, which means the design needs to understand what you can send to the print subsystem without being built with it included (ie a set of protocol standards)
I'm amazed that they beat the financial crisis. We all know Windows costs a lot for licences and today's governments are very cash-strapped. So like Portugal that recently announced they would not pay for any more Windows upgrades, I'm surprised more countries aren't looking closely at Munich to see if they can reduce their deficits slightly by going this route too.
to take the last part only:
imagine the history function is a method that's called. Does it really need PGO to speed up a function call? Or even if the display routines store up all links in a page then send them to a history function which returns all the ones that have been visited to be re-coloured after the final display is calculated, really does that need PGO? If one takes .25 seconds to run, would any call optimisation really shave any meaningful time off that number? The history module would still be optimised within itself after all, and just because the code is in a module doesn't mean it would have to manage the history data totally separately at runtime.
That's what I'm talking about here, and it could well be that the rendering engine is so huge it screws the optimiser by itself, it doesn't mean that a large majority of the non-critical-path codebase could be modularised, ie all the stuff surrounding the page rendering.
Of course, if such a small modification does make a large performance hit, then you have to wonder what the code paths are like. A certain amount of simplifying might make a much larger difference in such a case.
all distribution tot he public is a chore:
either C/C++ where you build for each platform .NET where you only have 1 platform, but the user still has to download the required .net runtime and security patches.
or
or Java where the user has to download the required JRE
You're probably best off with the Linux model where the code is downloaded and compiled on whatever platform the user is running.
I think the error message thing is just for templates too, as they're notorious to make easy to read. I've seen error messages for java apps where you have a stack trace that tells you nothing useful, and .NET exceptions that boil down to a Win32 structured exception code. None can be described as perfect.
So, yes, quite right - no language, including C++, is perfect. The FQA doesn't take that view though, it assumes C++ is the worst thing ever and doesn't try to hide its lack of objectivity. He hasn't exactly done a FQA for VB6 or PHP now has he.
I suppose you could use a VM based language where the optimisation takes place at run time instead, but they're trying to do all that in one step where it affects the developer (who can take the need to have a super computer to build) rather than the user (who doesn't want to have a super computer to run it), or an interpreted language where you have to write a VM to run it to get any form of acceptable performance.
Or you write it in something like C++ and complain that the optimiser tool uses too much memory itself (or that the developers didn't modularise it sufficiently)
Either way, the FQA is a terrible rant with plenty of over-exaggerations, inconsistencies and 'bad things' that were obsoleted decades ago. Ignore it and hope it goes away, or read this for some rebuttals.
I don't suppose the huge web-display system needs to be split into pieces (though I imagine there's not that much overhead to it - you won't get any PGO, but that's not that big an optimisation step as you'll still get it inside each module).
What could be done is to split the rest of the browser into modules. The code that displays a window doesn't have to be in the same module as the code that figures out how to display the data stream. The bookmarks system could be a separate module that's linked in at the end - its hardly essential that something like that is optimised to the operation of the whole.
disagree (and I'm not an Apple fanboi BTW). I think people here hate Apple becuase of their closed-in attitude. You buy an iPod and you only (effectively) can access it with approved Apple stuff. You buy a iPhone and you're locked into their store, etc etc etc.
We have to give Apple credit for kickstarting the whole smartphone industry and changing the world. You have to give them credit for popularising GUI interfaces and similar.
You also have to criticise them for the lock in and overpricing though, but their products aren't garbage (or no-one would keep buying them).
Naturally, bringing up Office means any UI arguments are automatically invalidated :) Office did things differently to everyone else for some reason, and everyone accepted and grumbled about it. But all the other apps were of a consistent standard. This was cited as part of Windows' lower TCO, and there was a book on how to make your application conform to the Windows UI style guidelines.
I think today's UI styles are back in the cowboy era of web design where you can do anything anywhere you like, and so designers do exactly that to differentiate themselves. I also think that they'll start to conform again when people have got used to the new stuff, or get so annoyed with geocities-style interfaces.
which is exactly why the new UIs are so poor.
Remember when Windows first came out, it has menus and every app had the same menu bar. Everything had a file menu that had new/open/print/etc items on it.
You could open any app and instantly know how to create a new document - because there was the file|new menu item, every time. You'd received training for all apps, instant familiarity, instant productivity.
Fast forward to today and we have different interfaces for everything. The new UIs with shiny orbs and animated transitions mean you have to figure out where all the new bits are for each app. Then some of them start working differently (eg Excel that has multiple icons in the task bar, but they're all running in a single instance so you close 1 you close them all kind of bo**ocks), and some don't even have menus - well, they have menus, but they're tucked away behind a little coloured icon so they appear when you click it, if you can find the f***er in the first place (eg the new hide-everything-away browser interfaces). The the ribbon comes along (which is a fine toolbar repacement BTW) but is used as a menu replacement too - with loads of bits hidden away in little menus behind tiny ">" icons.
The old interfaces were fugly, but functional. They made us productive and really that's what is needed for line-of-business apps. No-one really cares that excel looks cool, not when you're typing in the accounts.
sure, but MS supports all end-of-life products for 10 years, even VB6 was supported until recently.
What you get for your support however, generally means security bugs are squashed and MS will spend a tiny amount of effort making your apps work on future versions of Windows, if they feel like it. If you have a problem they'll tell you to install the latest service pack and if that doesn't fix it, tough - it doesn't support what you want it to do.
yep, Epson has one too, though it won't print pdfs for some reason. I guess because they expect you to just use it to send images to your printer. However, I use a WP4545 wireless printer and you just send whatever you like to it, from anywhere.
Epson also comes with a email printing service - you send your document to your own, unique email address and it forwards it to your printer. I don't have it set up, but it seems a nice idea.