Apple for a long time has represented 80+% of all the profits in the PC hardware business. They are now about 90%. Apple's marketshare is huge among consumers who aren't buying the cheapest systems. For example if you isolate to $2000+ laptops Apple's share is much larger.
It is just not true. Lets assume Apple grosses about $600 / iPhone. That would represent $2k in media sales or $1k / yr / iPhone customer in iPhone only media sales. Nowhere even close.
You do if you are selling in a high end niche. Apple is not dominating all phones but smart phones. Android has already crossed over to feature phones and probably around 2013 crosses over to dumb phones in real numbers. At that point in terms of phones Apple being competitive is over. But in terms of smart phones it could be even more dominant. It all depends on what the global percentage of smart phone sales are.
Europe is extremely heavily subsidized phone market. USA is moving in that direction.
Actually the fact that software won't run without a provisioning file, and that the phone doesn't not how to create a provisioning file for itself is a major technical advantage of iOS. The walled garder makes that technical advantage possible but the technical advantage is real.
30 years ago it was working great. It was still working pretty well 25 years ago. But Apple was not able to keep the spread in terms of cost or features reasonable.
Where Apple differs today is that the quality is much better, almost across the board and the spreads while present are not 2-3x but more like 15%.
Darwin is not a FreeBSD core. NeXTStep was a form of BSD, a cousin to SunOS and FreeBSD. Darwin has taken software from the wider BSD community and the BSD community has taken ideas from Darwin. FreeBSD and Darwin are close. But they are cousins, not father/son.
Second Linux is not just a kernel in common speech but a family of related operating systems. That was the whole point of the GNU/Linux naming convention, that Linux was being used widely to mean more than just a kernel.
Finally, the orignal poster never said anything of the sort. But the number of Windows developers dwarfs all the Unix/Open Source and Apple developers put together multiple times over. ASP and VB applications employed many tens of millions of man years.
Microsoft spends more on developing their OS than the rest of the world combined way more than Apple, it isn't remotely close. I can understand liking the Apple OS better, I'm an Apple user, but lets not lie about the situation.
As for installing apps, on Apple you should be using something like macports, fink or brew not raw compiles. In terms of software availability while ports is a so/so BSDish type system and Fink a stripped down Debian... the real Debian is a ton larger. Linux has Mac beat hands down. That's besides how powerful something like yum is, for which there is no Darwin equivalent. Again I love port, I use it all the time, but it pales next to most Linuxes. And raw source... blech.
No question. The iPhone as well. Prior to about 18 months ago the 80+% of the enterprise market in smartphones was Blackberry. They still probably have 40+% of the enterprise market even with their marketshare falling off a cliff. Apple and Android still lack vital enterprise features. I did an enterprise purchase about 2 years ago, and I gotta tell you they are missing stuff which I would have considered extremely important. I might have been able to get around it, but I'd love a much better Blackberry for an enterprise phone (and I use an iPhone for home).
Actually the dumb phone market is extremely fragmented as is the feature phone market. iPhone does in fact outsell every other phone on the market including dumb phones. Now if you rephrase this a bit like sales of iOS vs. JavaVM then things look different.
There are a large number of people who get their phones from their enterprise. Obviously enterprise features matter to them. Further many of the higher net worth people who were buying smart phones when they were only about 10% of the market consider exchange support a must have. During those days Android not having the feature was a break feature.
Now that isn't 95% of the market, nor is it 5% of the market. It is somewhere in between.
Honestly, in some ways I see it as a downgrade from what existed in 1995 or even 1991. I will agree the OLE is awesome. But on many of the core features I see stuff removed. For example look at the the list of functions in Excel today compared to what existed then. In 1993 Excel was talking about buying a mini version of the mathematica engine to do symbolic math, today user functions are even allowed, and they have a great strategy at research.microsoft.com. Access isn't standard anymore. I don't see why visio and project aren't bundled in. The note taking app is nice.
You are absolutely right though the products have been stagnant for 15+ years. And as a result lots of other products have caught up, in particular open office. That I think is what got Microsoft off their butts to finally push things along and resume being way ahead of the competition.
I don't think it is deranged to believe that office needs a new interface to grow. The easiest way to grow office is to essentially bundle in hundreds of "extensions" per product and allow each of those to fill a niche for each user. For example time value of money solver, IRR, NPV... solvers. Allowing for symbolic functions and user defined functions. An advanced stats package. Calculus and numerical methods. Powerful integration with dynamics. Etc.. Do a few hundred of those and suddenly everyone has one or two features that justifies the few hundred for Excel over Open Office. But.... to do that you need at least another level maybe two on hierarchical menus and that's too complex. So make it easier go with context based menus.
I want to see it to. So far Microsoft has great ideas but an inability to execute. But I agree the plan is a good one. I fully support the plan.
I don't buy it, contexts make sense. Power users of office just don't like the fact that their skills have been effectively diminished. It is the same reason that windows power users who do not develop are often the most negative about Linux desktops.
Tabbed toolbars which is what they were called in the 1990s have been around a long time. Contexts makes sense. I suspect Microsoft moving fast (for them) is more from the threat of open office. I think they wanted to be well ahead of open office for another decade or two.
My guess is they wanted to allow for a more gradual transition. Sort of like Windows 3.1 -> Windows NT.
They first shifted people over to the NT interface while leaving the DOS based GUTS the same same (Windows 95) then shifted away from DOS all together so stuff like PIFs just wouldn't work the same.
If 2020 office looks different and is different they want to move people gradually.
They can't have a version with a non functioning GUI so possibly the idea is
Step 1) Same functionality new GUI (change the GUI not the underlying product) Step 2) Additional functionality but same workflow Step 3) Additional functionality that requires a new workflow.
If you assume that's their goal. They don't want people having to go from Step 0 (legacy interface) to Step 2 directly. So they want to "force" people to learn ribbon.
I'm not so sure that paper so easy to use. Consider all the time spent teaching: manual writing, cursive, penmanship skills and then things like using a paper organizer. That is probably about 2 full years of education. I'm not sure that as a society it is worth it having people be skilled in using a pen / pencil anymore. I can see an argument for introducing laptop based methods early and never teaching those skills.
I will agree that after having invested all the time in paper competency paper seems easier.
My guess is that over time having these sorts of non context specific buttons won't work. The product is going to be able to get a lot more complex if all the interactions are specific.
Say for example you are doing the biography of a document. And you have things like the "footnotes" menu that will adjust footnotes in line with the biography, the link docs menu so that you can propagate biography changes into other documents that link or possibly all documents that use that reference (similar to linking OLE with a biography managers now)...
"paste values" is kinda vague. For example how should the system handle something like 23.45 in fixed format going into a float field? Maybe there should be an entire paste menu "paste values preserving formats", "paste values casting formats", "paste values asking me manually about format conflicts", etc...
Part of the issue is that while the ribbon change allows for this sort of layering, Microsoft hasn't done it yet. The product looks like a more confusing version of the old product not a product with tens of thousands of menu items making heavy use of AI determined contexts.
Also they added a virtualized directory system that the kernel shows applications. So applications that require DLLs in static locations with generic names get their version of the DLL linked in. I agree the name solution is far far easier and it was a flawed design choice that led to this problem.
I'm reading your comment. You do realize that Office for Mac 2011 has a partially ribbon interface? Anyway, according to Microsoft they are going to make the ribbon much more context sensitive by say 2020 the ribbon interface might represent 30,000 menu items.
The direction on Mac has been towards the ribbon interface. The old interface for Office for Mac was pre-existing. They aren't willing to develop both interface, rather what they have shown is a willingness not to discontinue them both at the same time. Given how much work they were doing for Office-Mac-2008 I can understand ribbon not being on the todo list.
Good argument, but no. If an upgrade deliberately destroys data that's not a bug only if it accidentally does it. Raskin's classic example of field blanking forcing someone to type something in again is very different from an upgrade that deletes a column in a database intentionally. I think the distinction is really subtle, using Raskin's analogy of the blanking field:
a) If I'm doing it because I didn't bother to cache the data, that's a bug. b) If I'm doing it because I believe it is important the end user retype, that's design.
Even though the end user experiences the same thing. A bug implies the software is not working as designed, arguably this change in design means that Apple believes the original was a bug, they never should have offered tentative / confirmed.
No but I know what collection of Desktop environments you are using. You aren't using Aero or Aqua. You aren't using one of the proprietary Unix GUIs like CDE or Display Postscript. It is for that reason that many people aren't clear what Android is a "Linux" in the proper sense since it doesn't use most of the Linux tool chain.
While there were people on both sides of the GNU/Linux debate (http://en.wikipedia.org/wiki/GNU/Linux_naming_controversy) the reason there was a debate was because the word Linux is being used to refer to an entire operating system.
OSX has pretty good handling for 2 monitor setups. I've used OSX that way and it is very nice, I love the ability to mirror and present. and I've used that all the time. I love the fact I can manipulate how the menu bar behaves. But that whole system is predicated on 2, and asymmetrical usage. I don't think the system was designed to handle 3 well.
Slashdot Mirror
Apple for a long time has represented 80+% of all the profits in the PC hardware business. They are now about 90%. Apple's marketshare is huge among consumers who aren't buying the cheapest systems. For example if you isolate to $2000+ laptops Apple's share is much larger.
It is just not true. Lets assume Apple grosses about $600 / iPhone. That would represent $2k in media sales or $1k / yr / iPhone customer in iPhone only media sales. Nowhere even close.
You do if you are selling in a high end niche. Apple is not dominating all phones but smart phones. Android has already crossed over to feature phones and probably around 2013 crosses over to dumb phones in real numbers. At that point in terms of phones Apple being competitive is over. But in terms of smart phones it could be even more dominant. It all depends on what the global percentage of smart phone sales are.
Europe is extremely heavily subsidized phone market. USA is moving in that direction.
Those exist. Some of the languages are very high level. Corona and livecode are good examples
Actually the fact that software won't run without a provisioning file, and that the phone doesn't not how to create a provisioning file for itself is a major technical advantage of iOS. The walled garder makes that technical advantage possible but the technical advantage is real.
30 years ago it was working great. It was still working pretty well 25 years ago. But Apple was not able to keep the spread in terms of cost or features reasonable.
Where Apple differs today is that the quality is much better, almost across the board and the spreads while present are not 2-3x but more like 15%.
While you are correcting him...
Darwin is not a FreeBSD core. NeXTStep was a form of BSD, a cousin to SunOS and FreeBSD. Darwin has taken software from the wider BSD community and the BSD community has taken ideas from Darwin. FreeBSD and Darwin are close. But they are cousins, not father/son.
Second Linux is not just a kernel in common speech but a family of related operating systems. That was the whole point of the GNU/Linux naming convention, that Linux was being used widely to mean more than just a kernel.
Finally, the orignal poster never said anything of the sort. But the number of Windows developers dwarfs all the Unix/Open Source and Apple developers put together multiple times over. ASP and VB applications employed many tens of millions of man years.
Microsoft spends more on developing their OS than the rest of the world combined way more than Apple, it isn't remotely close. I can understand liking the Apple OS better, I'm an Apple user, but lets not lie about the situation.
As for installing apps, on Apple you should be using something like macports, fink or brew not raw compiles. In terms of software availability while ports is a so/so BSDish type system and Fink a stripped down Debian... the real Debian is a ton larger. Linux has Mac beat hands down. That's besides how powerful something like yum is, for which there is no Darwin equivalent. Again I love port, I use it all the time, but it pales next to most Linuxes. And raw source... blech.
One reason. The iOS marketplace in dollar terms is 7x the size of the Android, Blackberry and Nokia marketplaces combined.
No question. The iPhone as well. Prior to about 18 months ago the 80+% of the enterprise market in smartphones was Blackberry. They still probably have 40+% of the enterprise market even with their marketshare falling off a cliff. Apple and Android still lack vital enterprise features. I did an enterprise purchase about 2 years ago, and I gotta tell you they are missing stuff which I would have considered extremely important. I might have been able to get around it, but I'd love a much better Blackberry for an enterprise phone (and I use an iPhone for home).
Actually the dumb phone market is extremely fragmented as is the feature phone market. iPhone does in fact outsell every other phone on the market including dumb phones. Now if you rephrase this a bit like sales of iOS vs. JavaVM then things look different.
There are a large number of people who get their phones from their enterprise. Obviously enterprise features matter to them. Further many of the higher net worth people who were buying smart phones when they were only about 10% of the market consider exchange support a must have. During those days Android not having the feature was a break feature.
Now that isn't 95% of the market, nor is it 5% of the market. It is somewhere in between.
You should have seen Oracle's pricing in the mid-late 1990s. I always feel like comparatively they are giving it away.
Honestly, in some ways I see it as a downgrade from what existed in 1995 or even 1991. I will agree the OLE is awesome. But on many of the core features I see stuff removed. For example look at the the list of functions in Excel today compared to what existed then. In 1993 Excel was talking about buying a mini version of the mathematica engine to do symbolic math, today user functions are even allowed, and they have a great strategy at research.microsoft.com. Access isn't standard anymore. I don't see why visio and project aren't bundled in. The note taking app is nice.
You are absolutely right though the products have been stagnant for 15+ years. And as a result lots of other products have caught up, in particular open office. That I think is what got Microsoft off their butts to finally push things along and resume being way ahead of the competition.
I don't think it is deranged to believe that office needs a new interface to grow. The easiest way to grow office is to essentially bundle in hundreds of "extensions" per product and allow each of those to fill a niche for each user. For example time value of money solver, IRR, NPV ... solvers. Allowing for symbolic functions and user defined functions. An advanced stats package. Calculus and numerical methods. Powerful integration with dynamics. Etc.. Do a few hundred of those and suddenly everyone has one or two features that justifies the few hundred for Excel over Open Office. But.... to do that you need at least another level maybe two on hierarchical menus and that's too complex. So make it easier go with context based menus.
I want to see it to. So far Microsoft has great ideas but an inability to execute. But I agree the plan is a good one. I fully support the plan.
I don't buy it, contexts make sense. Power users of office just don't like the fact that their skills have been effectively diminished. It is the same reason that windows power users who do not develop are often the most negative about Linux desktops.
Tabbed toolbars which is what they were called in the 1990s have been around a long time. Contexts makes sense. I suspect Microsoft moving fast (for them) is more from the threat of open office. I think they wanted to be well ahead of open office for another decade or two.
My guess is they wanted to allow for a more gradual transition. Sort of like Windows 3.1 -> Windows NT.
They first shifted people over to the NT interface while leaving the DOS based GUTS the same same (Windows 95) then shifted away from DOS all together so stuff like PIFs just wouldn't work the same.
If 2020 office looks different and is different they want to move people gradually.
They can't have a version with a non functioning GUI so possibly the idea is
Step 1) Same functionality new GUI (change the GUI not the underlying product)
Step 2) Additional functionality but same workflow
Step 3) Additional functionality that requires a new workflow.
If you assume that's their goal. They don't want people having to go from Step 0 (legacy interface) to Step 2 directly. So they want to "force" people to learn ribbon.
I'm not so sure that paper so easy to use. Consider all the time spent teaching: manual writing, cursive, penmanship skills and then things like using a paper organizer. That is probably about 2 full years of education. I'm not sure that as a society it is worth it having people be skilled in using a pen / pencil anymore. I can see an argument for introducing laptop based methods early and never teaching those skills.
I will agree that after having invested all the time in paper competency paper seems easier.
My guess is that over time having these sorts of non context specific buttons won't work. The product is going to be able to get a lot more complex if all the interactions are specific.
Say for example you are doing the biography of a document. And you have things like the "footnotes" menu that will adjust footnotes in line with the biography, the link docs menu so that you can propagate biography changes into other documents that link or possibly all documents that use that reference (similar to linking OLE with a biography managers now)...
"paste values" is kinda vague. For example how should the system handle something like 23.45 in fixed format going into a float field? Maybe there should be an entire paste menu "paste values preserving formats", "paste values casting formats", "paste values asking me manually about format conflicts", etc...
Part of the issue is that while the ribbon change allows for this sort of layering, Microsoft hasn't done it yet. The product looks like a more confusing version of the old product not a product with tens of thousands of menu items making heavy use of AI determined contexts.
Also they added a virtualized directory system that the kernel shows applications. So applications that require DLLs in static locations with generic names get their version of the DLL linked in. I agree the name solution is far far easier and it was a flawed design choice that led to this problem.
I'm reading your comment. You do realize that Office for Mac 2011 has a partially ribbon interface? Anyway, according to Microsoft they are going to make the ribbon much more context sensitive by say 2020 the ribbon interface might represent 30,000 menu items.
The direction on Mac has been towards the ribbon interface. The old interface for Office for Mac was pre-existing. They aren't willing to develop both interface, rather what they have shown is a willingness not to discontinue them both at the same time. Given how much work they were doing for Office-Mac-2008 I can understand ribbon not being on the todo list.
Good argument, but no. If an upgrade deliberately destroys data that's not a bug only if it accidentally does it. Raskin's classic example of field blanking forcing someone to type something in again is very different from an upgrade that deletes a column in a database intentionally. I think the distinction is really subtle, using Raskin's analogy of the blanking field:
a) If I'm doing it because I didn't bother to cache the data, that's a bug.
b) If I'm doing it because I believe it is important the end user retype, that's design.
Even though the end user experiences the same thing. A bug implies the software is not working as designed, arguably this change in design means that Apple believes the original was a bug, they never should have offered tentative / confirmed.
No but I know what collection of Desktop environments you are using. You aren't using Aero or Aqua. You aren't using one of the proprietary Unix GUIs like CDE or Display Postscript. It is for that reason that many people aren't clear what Android is a "Linux" in the proper sense since it doesn't use most of the Linux tool chain.
While there were people on both sides of the GNU/Linux debate (http://en.wikipedia.org/wiki/GNU/Linux_naming_controversy) the reason there was a debate was because the word Linux is being used to refer to an entire operating system.
OSX has pretty good handling for 2 monitor setups. I've used OSX that way and it is very nice, I love the ability to mirror and present. and I've used that all the time. I love the fact I can manipulate how the menu bar behaves. But that whole system is predicated on 2, and asymmetrical usage. I don't think the system was designed to handle 3 well.
More than 2 is just highly uncommon.
The code base for the Mac and Windows products have almost nothing in common. Don't judge the product direction by Mac.