Comments

Actually, cmd.exe in Windows XP has tab-completion.

The only thing it's still missing compared to bash in that regard is that it won't work on files in your path, only in the current directory. I wonder if this fixes it. I can't check, because I can neither install the preview nor watch the video, as my computer is acting really really strange. I'm currently working in Knoppix, and it's gone for about 25 minutes now without crashing, which is the best it's done all day. In fact, the previous three attempts in Knoppix all crashed within 30 seconds. At least Windows consistently gives me 3 minutes before giving up.

Looks like a hardware defect, yes, although I can't explain why it's suddenly gone away it seems. If it lasts till the half hour I'm going to boot Windows back up and try my luck with that one more time.

This sounds really, really interesting.

In the middle of the video, before I would forget to do so, I ran off to BetaPlace (whose URL is not betaplace.com, but http://beta.microsoft.com, and has been for quite some time now) and filled in the survey.

I'm very curious on how this'll turn out.

dotnetjunkie wrote:

I am a VB.NET guy and I care A LOT about javascript, it's very important to create smooth, user-friendly web apps.

Ditto.

Agreed. Express should be free. It's free now, it should still be free come RTM. Just stick a "not for commercial usage" EULA in there and it won't cost MS any money.

In fact, it'd probably make them money. If we have Joe Hobby-Programmer who uses VB.NET Express at home (for free), and he gets hired to do some project for a company, naturally he'll want to stick with what he knows, at which point a full license for VS2005 will be needed.

If MS is going to charge money for Express, it's a big mistake. It doesn't matter is they charge $100 or 10 cents, just the fact that you have to pay anything, regardless of how little, scares off a lot of people, especially teenagers who don't have a credit card (which I suspect is a big target audience). I started programming when I was 10. If GW-BASIC had not come with DOS back then, if I had to pay even one old-fashioned Guilder for that, I probably wouldn't have done it.

Express should, no, must be free!

Care to elaborate on that? What's so bad about VB.NET? I've used MBASIC, MSX-BASIC, GW-BASIC, QuickBASIC, PowerBasic and all incarnations of Visual Basic from version 3.0 up, and I absolutely love VB.NET.

Of course personal opinions vary and I'm not questioning your right to hate VB.NET, but I'm curious what is it that makes it so bad for you.

I do use C# sometimes, but basically only when I need either unsafe code or unsigned types.

Manip wrote:

I would describe an intuitive menu as a 'smart menu' that tries to guess what the user to likely to be doing and helps that user complete it. So instead of providing a raw list of all possible items it provides common tasks that make it a lot more user-friendly.

An example of this is the Microsoft personalised menus. As it learns what you use commonly it will start stripping out all unused items and thus your menus become more intuitive.

I disagree. Although for the Start Menu the personalised menu worked quite admirably (and the Start Panel works even better), for Office I think the personalised menus are the worst thing ever to have happened to it. Menus that reshuffle themself will screw up your spatial memory, and spatial memory is an important aid in being able to use any device or piece of software with any kind of speed. I for one am really, really glad that this is one piece of the Office UI that never made it into Visual Studio. Of course, this remains an opinion, and ymmv.

I like the idea of the "horizontal parent menu" that is shown here. Indeed it'll allow you to easily browse other albums/artists etc. while viewing information at a greater detail.

If you're so intent on smart menus, I'd prefer them the way WMP's Auto Playlists work - i.e. using additional lists for often played or highly rated content. I wouldn't like it to go around hiding stuff in the normal menus.

I once had to do some prototyping in Flash for the University course Human Computer Interaction (aka "Interfaces"), and even that was really really bad. Over the course of that project I have really come to loathe Flash MX and its scripting language. I guess I've been spoiled with Visual Studio et al.

I think there's only one scripting language in the world that I hate more than the Flash scripting language... and that's InstallShield's InstallScript.

Easier, perhaps. Less readable and more error-prone, definitely.

It's probably a matter of taste, but I prefer the C++ way whenever there is one anyway.

#include <iostream>
#include <string>
#include <algorithm>
using namespace std;

bool IsPalindrome(string sz)
{
  string szReverse(sz.length(), ' ');
  reverse_copy(sz.begin(), sz.end(), szReverse.begin());
  return szReverse == sz;
}

bool IsPalindrome2(string sz)
{
  string::iterator forward = sz.begin(), backward = sz.end() - 1;
  while( forward < backward && *forward == *backward ) {
    ++forward; --backward;
  }
  return forward >= backward;
}

template<typename T>
bool IsPalindrome3(T sz)
{
  T::iterator forward = sz.begin(), backward = sz.end() - 1;
  while( !(forward - 1 == backward || forward == backward) && *forward == *backward ) {
    ++forward; --backward;
  }
  return (forward - 1 == backward || forward == backward);
}

int main()
{
  string sz;
  cout << "Please enter a string:" << endl;
  getline(cin, sz);
  if( IsPalindrome2(sz) )
    cout << "It's a palindrome!" << endl;
  else
    cout << "It's not a palindrome!" << endl;
 
  return 0;
}

Three versions in C++, the first one is the lazyman's version, using std::reverse_copy. The second one is the iterator implementation. And since I figured I was using iterators anyway, so why not make a templated version that works on any standard container that defines a begin() and end() that return a bidirectional iterator. That way you can check a std::list for being a palindrome too! ^_^
You could also make a version two template iterator parameters, like many of the <algorithm> functions, that would work on any bidirectional iterator, including pointers.

Sounds interesting. You'll defenitely need to get those secure C APIs standardised though, everybody from Linux to MacOS could benefit from that kind of thing. Not that they'd trust it if it comes from Microsoft, but still...
Also, real C++ programmers don't use C APIs anyway, I mean, who needs scanf and char* when you've got std::cin and std::string (both of which are safe btw)?

Also, it's not exactly clear to me what's so new about this stack buffer overrun protection... VS.NET 2003 does that too. Sure, I assume you've optimised it more, but is there anything really completely totally new added to it? That didn't become very clear from the video to me.