If you don’t know what SOPA is and you don’t feel like spending more than 3 minutes on the subject, watch this video.

Chrome can do things that no other browser can do, and Google now targets Chrome exclusively, meaning some Google sites only function fully when viewed in Chrome. Even today, you can read on the Google blog about some new Angry Birds levels that only work in Chrome. This is disturbing, when you consider that Google has made a lot of hay about the openness of the Web.

To read more: http://www.pcmag.com/article2/0,2817,2397158,00.asp

The Three Rules of Wireless Networking (source):

1. It never goes as fast as they say it does
2. It never goes as far as they say it does
3. It never sets up as easily as they say it does

Coming soon

“Then we have used the SCD to perform the No PIN attack at the local cafeteria and
even in some random shops in Cambridge. We have successfully bought books and DVDs
worth over $50 at one of the shops using the journalist’s card but typing PIN 0000. Even
more, we have performed the tests without warning and nobody has noticed the hidden
device or fake card (the card interface connected to the SCD). After the transaction we
have disclosed the attack to the shop manager who said that such attacks occur very often.
The manager mentioned that during busy periods like Christmas credit card frauds occur
at least once a week. Because shops cannot longer check the cards (as the current policy
is to let the customer handle the card insertion and removal) the criminals can use fake
cards and devices similar to the SCD to perform fraud.”

The entire thesis can be read here.

I recently tried to use Bitmap.scaleInto() to resize transparent bitmaps in my application. Although this function is a nice addition to JDE 5, it tends to produce bad-looking results on transparent bitmaps such as PNG24 files. Forum posts I found on the web for this problem talk about using Bitmap.createAlpha() which does restore transparency but still produces artifacts in the final bitmap such as white edges and wicked pixel colors.

To solve this without the use of an external library, I created the following function. It still uses scaleInto() but in a different manner. I added some comments in the function and created a screenshot showing before and after results (don’t mind the pictures, they’re part of the project I’m currently working on). I’m posting the result here in hope it will help other people searching to achieve the same thing. The code was tested in JDE 5 and JDE 6 and the sample pictures were created using different combinations of filter types and aspect ratios.

package com.patchou.ui;
import net.rim.device.api.system.Bitmap;
import net.rim.device.api.ui.Color;
import net.rim.device.api.ui.Graphics;
/**
 *
 * @author    Patchou
 * @version   1.01
 *
 */
public class GPATools
{
    /**
     * Resizes a bitmap with an alpha channel (transparency) without the artifacts introduced
     *   by <code>scaleInto()</code>.
     *
     * @param bmpSrc        Source Bitmap
     * @param nWidth        New Width
     * @param nHeight       New Height
     * @param nFilterType   Filter quality to use. Can be <code>Bitmap.FILTER_LANCZOS</code>,
     *                           <code>Bitmap.FILTER_BILINEAR</code> or
     *                           <code>Bitmap.FILTER_BOX</code>.
     * @param nAspectRatio  Specifies how the picture is resized. Can be
     *                           <code>Bitmap.SCALE_TO_FIT</code>,
     *                           <code>Bitmap.SCALE_TO_FILL</code> or
     *                           <code>Bitmap.SCALE_STRETCH</code>.
     * @return              The resized Bitmap in a new object.
     */
    public static Bitmap ResizeTransparentBitmap(Bitmap bmpSrc, int nWidth, int nHeight, int nFilterType, int nAspectRatio)
    {
        if(bmpsrc== null)
            return null;

        //Get the original dimensions of the bitmap
        int nOriginWidth = bmpSrc.getWidth();
        int nOriginHeight = bmpSrc.getHeight();
        if(nWidth == nOriginWidth && nHeight == nOriginHeight)
            return bmpSrc;

        //Prepare a drawing bitmap and graphic object
        Bitmap bmpOrigin = new Bitmap(nOriginWidth, nOriginHeight);
        Graphics graph = Graphics.create(bmpOrigin);

        //Create a line of transparent pixels for later use
        int[] aEmptyLine = new int[nWidth];
        for(int x = 0; x < nWidth; x++)
            aEmptyLine[x] = 0x00000000;
        //Create two scaled bitmaps
        Bitmap[] bmpScaled = new Bitmap[2];
        for(int i = 0; i < 2; i++)
        {
            //Draw the bitmap on a white background first, then on a black background
            graph.setColor((i == 0) ? Color.WHITE : Color.BLACK);
            graph.fillRect(0, 0, nOriginWidth, nOriginHeight);
            graph.drawBitmap(0, 0, nOriginWidth, nOriginHeight, bmpSrc, 0, 0);

            //Create a new bitmap with the desired size
            bmpScaled[i] = new Bitmap(nWidth, nHeight);
            if(nAspectRatio == Bitmap.SCALE_TO_FIT)
            {
                //Set the alpha channel of all pixels to 0 to ensure transparency is
                //applied around the picture, if needed by the transformation
                for(int y = 0; y < nHeight; y++)
                    bmpScaled[i].setARGB(aEmptyLine, 0, nWidth, 0, y, nWidth, 1);
            }

            //Scale the bitmap
            bmpOrigin.scaleInto(bmpScaled[i], nFilterType, nAspectRatio);
        }

        //Prepare objects for final iteration
        Bitmap bmpFinal = bmpScaled[0];
        int[][] aPixelLine = new int[2][nWidth];

        //Iterate every line of the two scaled bitmaps
        for(int y = 0; y < nHeight; y++)
        {
            bmpScaled[0].getARGB(aPixelLine[0], 0, nWidth, 0, y, nWidth, 1);
            bmpScaled[1].getARGB(aPixelLine[1], 0, nWidth, 0, y, nWidth, 1);

            //Check every pixel one by one
            for(int x = 0; x < nWidth; x++)
            {
                //If the pixel was untouched (alpha channel still at 0), keep it transparent
                if(((aPixelLine[0][x] >> 24) & 0xff) == 0)
                    aPixelLine[0][x] = 0x00000000;
                else
                {
                    //Compute the alpha value based on the difference of intensity
                    //in the red channel
                    int nAlpha = ((aPixelLine[1][x] >> 16) & 0xff) -
                                    ((aPixelLine[0][x] >> 16) & 0xff) + 255;
                    if(nAlpha == 0)
                        aPixelLine[0][x] = 0x00000000; //Completely transparent
                    else if(nAlpha >= 255)
                        aPixelLine[0][x] |= 0xff000000; //Completely opaque
                    else
                    {
                        //Compute the value of the each channel one by one
                        int nRed = ((aPixelLine[0][x] >> 16 ) & 0xff);
                        int nGreen = ((aPixelLine[0][x] >> 8 ) & 0xff);
                        int nBlue = (aPixelLine[0][x] & 0xff);

                        nRed = (int)(255 + (255.0 * ((double)(nRed-255)/(double)nAlpha)));
                        nGreen = (int)(255 + (255.0 * ((double)(nGreen-255)/(double)nAlpha)));
                        nBlue = (int)(255 + (255.0 * ((double)(nBlue-255)/(double)nAlpha)));

                        if(nRed < 0) nRed = 0;
                        if(nGreen < 0) nGreen = 0;
                        if(nBlue < 0) nBlue = 0;
                        aPixelLine[0][x] = nBlue | (nGreen<<8) | (nRed<<16) | (nAlpha<<24);
                    }
                }
            }

            //Change the pixels of this line to their final value
            bmpFinal.setARGB(aPixelLine[0], 0, nWidth, 0, y, nWidth, 1);
        }
        return bmpFinal;
    }
} 

Here’s an example of how to call the function:

Bitmap bmp = Bitmap.getBitmapResource("picture.png");
Bitmap bmpResized = GPATools.ResizeTransparentBitmap(bmp, 30, 60,
    Bitmap.FILTER_LANCZOS, Bitmap.SCALE_TO_FIT);

Here’s the result:

Today, something happened that will help shed some light on my no-beta-support policy. As some of you already know, Microsoft recently released a beta version of Windows Live Messenger “Wave 4″. This updated version of Messenger includes new social features, options like tabbed chats, noticebly larger advertisements, etc… Microsoft also modified their setup program, that same setup I downloaded this morning and ran on my computer.

After a quick installation, I was invited to restart my computer, which I did, promptly. This is where things started to get interesting to say the least. After the reboot, I was greeted by brand new error messages I had never seen before in Windows, one in particular saying that my profile, now located in “system32\config\systemprofile”, was not available. Basically, my whole user profile was gone, my desktop empty, the taskbar displayed default blank icons for my old shortcuts and none of my programs remembered who I was. Messenger’s new signin screen was nagging me, coming back after I closed it, and many of the things I tried to do in Windows while trying to figure out what happened displayed error messages. In user terms, my computer was completely “broken”.

After searching a bit about this issue on the web, the only thing I was able to find was posts confirming that this kind of error occurs when a user profile has been badly corrupted in the system and that, in such cases, there’s generally nothing to do but create a brand new profile. For all the tech geeks reading this, yes, I checked the user profile keys in the registry, I tried to clear up the default system profile, etc… I also don’t have anything out of the ordinary running on my station, no shell replacement software, no theming program, no nothing: just Windows 7, Office 2010, Visual Studio and a couple of work tools. I don’t even play games on that system.

Out of despair, I tried System Restore and I have to say it performed admirably. It finally reverted all the changes that the new Windows Live setup had done to my system and restored my user profile to a functionning state. I’ll definitively have to remember to rely more on this feature in the future. In any case, as I refused to admit defeat, I prepped my computer for another try. I erased all the files and settings of my existing version of Messenger (the current public release), did a clean reboot, re-downloaded the setup of the beta version and tried again. Guess what: after the restart, the same error occured, my profile was corrupted again and my computer was back to stone age.

Now, I hope you understand why supporting a beta may not be a good idea. As for Microsoft’s work on this new Messenger, I can say I’m not impressed, at all. I’ve never seen a program cause so many issues on my system and I certainely don’t expect that from an instant-messaging application. Beta or not, I assume that when Microsoft releases something to the public, it’s not going to cause mayhem. I don’t know what that new Messenger is trying to do on our systems but in my view, it’s wrong and uncalled for. Here’s a recommendation for everybody who’s reading this post is: stay away from Messenger beta versions, I certainely will.

In the last 2 days alone, I read the following two articles about Apple and how they treat their developers and users. To sum it up: you can’t choose the technology you’ll work with, you can’t choose the features you’ll add and now, you can’t even choose the tools you’ll use for your development. Next time you hear about Microsoft or another software company being sued in Europe, think about them. As for me, I’ll continue to create software with a company who respects and takes care of its developers.

http://www.neowin.net/news/apple-rejects-app-for-using-pinch-to-expand-gesture

http://www.neowin.net/news/apple-bans-adobe039s-flash-to-iphone-compiler-plus-related-applications

Today, I feel like talking about technology. While selecting the hardware for an office my company is opening in Montreal, I came across an interesting issue regarding motherboard. The P7Q57-M DO is a nice product, from ASUS. It features one of the latest chipsets from Intel (the Q57 Express), lots of cool Intel technologies for businesses, the latest CPUs, etc… go ahead, check-out the links. Heck, it’s even “ready for Windows7″! incredible stuff I tell you.

If you check the picture of the motherboard, you’ll notice it has a bunch of graphic ports: a VGA port, a DVI port and even an HDMI port. Asus’ web site mentions the max resolution for each one, bpp and even frequency. All that for 135$ CAD at Newegg, pretty good deal right? well, maybe, but there’s a catch: if you use one of those cool Core i7 processors Intel has been distributing recently, those graphic ports serve no purpose. The fact is, the new H55/57 line of chipsets from Intel needs a graphic unit embedded in the main CPU to output video. Unfortunately, this tiny detail seems to be mentioned nowhere on Intel’s Q57 chipset page or on Asus’ web site. I bought two of those cards and I can tell you this information is nowhere to be found in the manual either.

Surprisingly, there seems to be no review of this motherboard on hardware specialized web sites, and this is why I felt I had to write this post. I hope it will prevent some of you to make the same mistake I did when I bought those motherboards. If you buy one and plan on using the integrated graphic ports, checkout this support page on Intel’s web site to see which CPU will give you video. As for me, I’ll stick to an i7 and I’ll just combine it with an ATI 5450 (50$, twice as powerful as Intel’s integrated solution and it does not even require a fan). The end result is nice, I just wish Intel and especially Asus had done their homework when publishing their specs and manual.

This time, it’s all coming from a post from Eric Ligman on MSDN. Unfortunately, it seems that many people just misunderstood what the article is saying. I’m quoting: “technically possible does not always mean legal”. In his post, Eric does not point a finger at all the people who’re tricking the setup of Windows 7 to install onto a freshly formatted hard drive. His post talks about licenses, and licenses only: if you buy an upgrade license of Windows 7, for it to be valid, you need to own fully valid license of Windows XP or Windows Vista. That’s it, end of the discussion, that’s why upgrades are priced differently from full products. He does NOT imply that using the installation hack breaks your license to use Windows as it does not.

The problem at hand is that people always mix technical things and legality issues. This case is similar to when people were saying that you could change a couple of registry keys in Windows 2000 to make it work like the Server Edition. That was technically possible, but legally, you did not ended-up with a Windows Server license so people using this trick were probably be better off just pirating the real CD of Windows Server in the first place. Eric’s post is directed toward all the general users who may be in search for a bargain for their Windows 7 shopping. Some people would find this way to hack the installation, would not understand the possible legal requirements (= owning a license of XP or Vista) and would happily buy an upgrade of something they don’t have. Those people would not be thinking they are pirating anything and these are the same people who could complain later on if WGA bugs them about the issue (it won’t happen in this case but you see the point).

If I buy myself a GPS and set of brand new tires, I won’t expect Audi to give me a car to go with them, even if that GPS and those tires are designed specifically for Audi cars. The same logic applies to software and software upgrades. In the end, it’s just a matter of paying for what you’re using. I agree Microsoft probably made a bad decision in preventing people from doing fresh installs like they could before, I also find that annoying, but this is not an excuse for pirating it. If you own an XP or a Vista license, you’re free to install Windows 7 anyway you like, really, I don’ t see why Microsoft would mind. And to those claiming Microsoft is ripping them off, I have a solution for you.