I’v been using Windows 7 for a couple days now and its waaaaaaaaay better than Vista. It boots faster and is snapper, then XP. Also they fixed UAC. It only took up about 5G of space and the iso was only 2G.
Plus I tried it in a VM with only 512ram and it ran fine.
I think they got it right this time. Can’t wait until it comes out.
So, by this logic, Linux is dead?
Most people still using IE 6 either:
a) Have no clue how or why to upgrade to IE 7
b) Are on an old OS that doesn’t support IE 7
c) Are working in a large org with an IT department that doesn’t support upgrading to IE 7
Out of these three, the most IE 6 users are because of a combination of b) and c).
Applications loading more quickly holds no immediate interest for them; they are still ‘waiting for something to happen’.
It really depends how much quicker the process runs. If a process-intensive task goes from 18 minutes to 17 minutes, users won’t notice, since they’ve probably gone off to do something else anyway. But if the improvement is from 61 seconds to 1 second, people will notice. Even a change from 15 seconds to 5 seconds will be noticed. See Jakob Nielsen’s article [http://www.useit.com/papers/responsetime.html] on the topic. It’s a bit old, but people’s reaction times probably haven’t changed much over the last 3,000 years or so.
That screenshot doesn’t inspire me with admiration for the fit and finish of Windows 7. There’s a button that (apparently) doesn’t do anything! The logical shifts have left on the left, right on the right (duh) but the rotates have left on the right and right on the left! They’ve interleaved shifts with bitwise operations for no imaginable reason! They’ve used the same (big, monospaced) font for showing bit positions as for the bits themselves! These are all little quibbly things, but what they show is that whoever implemented these things doesn’t care about getting obvious things right. And: if I’m correctly interpreting the three different styles of button, the memory operations are disabled in programmer mode. That’s pretty poor. And: why should square root be disabled, when division is allowed?
So yes, OK, I’m sure it’s an improvement on the previous version. But there are a bunch of things that are wrong even though doing them right would have been just as easy, and a bunch of other things that are wrong even though doing then right wouldn’t have been much harder. Fair enough, I guess, since the calculator isn’t all that important a part of the system, but this is not the sort of thing that gives me confidence in the fit and finish of the system.
… that for example computing 10.21 - 10.2 resulted in 0.0100000000000016. Today, Calc’s internal computations are done with infinite precision for basic operations (addition, subtraction, multiplication, division) and 32 digits of precision for advanced operations (square root, transcendental operators).
I wonder if M$ reinvented the wheel for this, or if they used GMP?
Probably not judging by the arbitrary restriction of 32 digits… No brainer I guess.
That calc ui is very web2.0. Guess that’s why users want these days, everything to look like a Mac?
What are talking about? How does it look very web 2.0?
I’m not sure the Chen post is quite right. Whatever massive improvements calc had in Vista it still pretty much did the same things as far as an everyday usage scenario is concerned. Users should not be expected to notice underlying changes that should have been right in the first place!
What the Win 7 calc does is add real world functions like gas mileage, wages, time duration and so on. That users notice and with good reason. It makes itself more useful not just more technically capable. Plus the template feature makes it easy to add new functions. You could do this in unix calculators of course but programming the calculator usually required using your own or another person’s shell script.
Maybe I’m just old-school - but I don’t replace something until it has out lived its usefulness.
I don’t replace my car until the cost of future repairs starts to mount (or the reliability drops) - which normally means a 10 year cycle between purchase and trade in.
But hardware doesn’t wear out quite the same way…
I don’t replace my computer until technology has moved on a couple of steps - normally I upgrade every 3 years or so (standard depreciation for computers by my tax system).
As for software - hystorically an upgrade was almost mandatory to use the latest gadgets and software. I have followed every MS DOS and Windows version from DOS 1 and Windows 3.1. I followed every upgrade (yes - I went down the Windows ME road block). I even went through NT and then to XP. But I stopped at XP and never went to Vista. Why? None of my hardware would work on it, and Vista didn’t offer enough for me to warrant throwing out the hardware and buying new stuff. Every other version of DOS and Windows had some degree of backwards compatiblity to software and hardware (unless you had lotus and tried to upgrade DOS… MS were more blatent in those days).
Yes I will move on to Windows 7 - probably next year. But the point is that Microsoft were their own worst enemy with Vista. They over hyped it (even the name was over hype) and then cut their own throats due to software and hardware incompatibilities. And as for look and feel/Polish - I’d rather be using my Mac Book Air with Leopard OS.
The damn problem with Windows’ interface isn’t the fact that the UI for Calc wasn’t updated for years. It’s just another example of the sloppiness that goes into Windows fit and finish.
I don’t really care too much about cool looking interfaces, but it is nice to know that everything works the same way and looks like it belongs together. When half of the dialog boxes still present a Windows 3.1 view of your computer, and the other half present a Windows 95 view of the world. When half of the Windows look one way and the other half the other way. When the whole OS looks like it was tossed together from part in the bargain bin, it affects the user’s experience of the OS.
Much of the Macintosh Finder code is pre-Mac OS X! Yet, Apple makes sure it looks and acts the same as the rest of the OS. They even have added improvements to it to make it a bit more useful. Only in the next OS release will the Finder be updated with all new OS X code.
If you ever look in the auto magazines, the biggest complaint against GM isn’t the dependability of their cars, or their technological backwardness, it’s their dashboard. The cheap plastic look, the poor fit and finish, the poor arrangement of gauges, and the fact that the dashboard still the same as the dash that was in the disco era models color consumer’s perception of the car.
Now, it isn’t that Toyota uses exotic woods in their dashboards, and I doubt that Toyota spends a lot more than GM in design and execution of their dash, but the better quality of Toyota dashboards gives the perception that Toyota simply cares more about their cars’ quality than GM does. After all, when you buy a car, you spend more time looking at the dash than the outside.
The funny thing is that Windows spends a heck of a lot more than Apple does in usability labs, has almost 100 times more programming resources, and yet for some strange reason, Apple still does a better job with fit and finish. Maybe this hasn’t been Microsoft’s high point. Maybe Microsoft gets so into various trends like Areo without thinking about the big picture.
I have an application on my iPod Touch called Air Sharing. It allows you to upload and download documents to your iPod Touch or iPhone, and uses WebDAV.
There are directions on Linux, Mac, and Windows. The directions for Mac is two step (Go to the Sharing menu, and type in the URL displayed in Air Sharing). Linux is similarly short although separate directions are given for Gnome and KDE.
Windows is another story, It’s a ten step series of instructions from opening the Start Menu to warning you that Vista might ask you for your password and name multiple times. Oh, yes I said Vista. There are separate directions for Windows XP starting with Patching Windows XP.
It isn’t about whether user’s notice or don’t notice changes, its the idea that fit and finish colors a user’s perception about quality. How long does it take to put a more colorful Windows Vista look to the Calculator? It probably is a simpler job than rewriting the whole guts, so why not simply slap on a new look while you’re at it?
Damn, I don’t give a hoot who steals what from who, but Microsoft should at least steal one thing from Apple’s playbook: UI determines how a user interacts with the application, and it isn’t about flash as much as consistency. In fact, Apple toned down the Aqua UI because it was too colorful and flashy. But, when Apple did tone down the UI to make it less splashy and more friendly, they did it across the board.
I have a lot of complaints about Windows, but underneath it is a damn solid OS. However, the UI is sloppy and poorly thought out.
To me linux never made sense on the desktop. It makes more sense on the server end. Especially if you start a company and the application needs to scale to a lot of servers.
The overall tone of people who can’t resist saying things like ‘I switch to Mac two years ago’ or ‘Thanks god I use Linux’ is really annoying.
I agree that unless you present your changes in something that that users can see, nobody will notice. Some people take Jeff statement to the extreme, drawing conclusions such as Don’t make quality software. Make software that sells., while I guess what Jeff trying to say is, even if you make some change and improvement under the hood, make sure it has some visual cues so the user will notice and appreciate it
Much of the Macintosh Finder code is pre-Mac OS X! Yet, Apple makes sure it looks and acts the same as the rest of the OS. They even have added improvements to it to make it a bit more useful. Only in the next OS release will the Finder be updated with all new OS X code.
Actually, the Finder was completely rewritten for Mac OS X. I think you’re mixing up Carbon with Cocoa. The Finder is written in Carbon, which is a holdover for applications that arn’t written in Cocoa (object-orentiated objective-c API).
I guess I don’t really get why increasing the precision of the Calculator is some massive improvement that should have impressed users. In circumstances where I care about that kind of precision, I don’t want to be using that crappy little app that imitates a physical calculator for no real reason. I want a proper command line tool, with visible history and the easy ability to call back and edit previous calculations.
So, I’m one of those programmers who late last year got a MacBook Pro instead of a laptop with Vista. My second day using it I accidentally discovered how to summon the calculator app with a single keystroke. (For those who don’t know – F12 or fn-F4 summons a simple calculator, the current weather and a five day forecast, a clock, and a calendar.) Maybe there’s a way of doing that on Windows, too, but I’ve never learned it in 18 years of daily Windows work. For my money, even though the Mac calculator is very basic, that ease of access to it is worth far more than the missing functions. If I need something more powerful, there are a wealth of Unix tools sitting at my (always open) terminal prompt.
stronghi/strong
Honestly, who cares if they rewrote the calc engine? I don’t. I’d vastly prefer if they added some additional basic functionality, like perhaps a graphing calculator mode. As an added bonus, Apple has had this function for 15 years, and Microsoft always likes to copy Apple on everything.
Vista was and is a turd sandwich. Doing things like removing the ability to go up a level in the Windows Explorer with a standard shortcut (up-arrow, and no, clicking on the tiny folder names isn’t a good substitute) and button (backspace) – which is quite possibly the most common folder navigation one uses besides going down into a folder – is just hideous UI design. You simply don’t want to take a common operation and remove it (backspace was changed to Back, which is not what you want half the time) or make it harder to use.
Common operations should be the easiest to use.
Maybe it’s because I’ve jumped around so much, but I’ve owned several macs and PCs over the last 5 years, when I switched to Vista, I had no trouble migrating from XP/OSX to it.
I think having a mindset that you dislike something because it’s different or got bad reviews is a sure-fire way to make sure you don’t have a good overall experience.
I’m not partial to any OS, I’ve used them all and appreciate them all. I currently use Vista because I’m doing .NET, it’s the most appropriate tool, and no I don’t have OS-Envy, I just use the tool I feel would suit me best at any given time.
I loaded it on an IBM T40 and was impressed. Lots to like…
Jeff, I too am a fan of HFI (one class away from my cert!). But I think you’ve missed the point of that particular button.
The idea isn’t to make sure that every updated feature or performance improvement is bubbled-up to the UI layer (yuck), but that features ought not be hidden or difficult to find. They should be exactly where the user expects them to be.
That said, I continue to applaud you for bringing usability concepts to the unwashed developer masses. 
If I don’t buy it, I don’t know the new functions are there. Why doesn’t Microsoft just update XP all the way? There is nothing wrong with XP except that Microsoft needs a new name for the os (Vista) and bundle it into a nice box that consumers can grab.
Choosing an OS based on the look of its calculator app is like choosing a wife based on the look of her tits.