I did my time.And I have more than made it up through my deeds.You insensitive pr***. To call me a criminal is rather ignorant and insulting, I did not referrence my lurid past to highlight any of my failings but rather to give you and any other readers here knowledge of where I am coming from, if you wish to insult rather than to debate then go ahead and do that, you just wont find an answer in return.
Very well, put your money where your mouth is. Please provide a reference that indicates that a larger user base of a program will cause more successful exploits. We await your facts.
If all my reasons explained before doesnt convince you nothing will. If you are so naive to believe that even if we had thousands of vendors out there adapting themselves to work with the differences in Firefox or Opera and that if you were a hacker or a insidious individual who may design programs to take advantage of your computer by selling mal-ware designed to increase awareness of your site by attacking not the least used programs but the most used programs, then I simply cant help you, I could probably go out and find articles on this or official documentation. You are arguing security I am arguing availability, your bull headedness on this issue is obviously corrupting your vision on this.
So I will try and make this one last attempt.
Have you ever tried counting cards at a black jack table? Card counting uses the law of large numbers, its not what people think it is when you count cards. You are simply guessing at what comes next in the count if say the dealer has dealt 5 cards under ten previously than it is likely that a ten will be dealt next. In the short run you are likely to come up even at a blackjack table doing this or maybe even slightly ahead, if you multiply the participants you increase your chances at a profit when you multiply that by numerous sessions over numerous days you will most likely beat the house with a 5% edge, which is huge in gambling terms. The same law applies to a largely distributed program.
An even simpler way is to say that the number one car in the world is a VW, a VW will probably have more numerical accidents than any other vehicle, forget its survivabilty rate or casualty rate, greater number=greater number of accidents.
That is simply all I have been saying. If you prefer to throw math out the window, which I highly doubt you will then it should stand.
Past behavior is indicative of future actions. If you've done your time and reformed then I'm happy for you, but as a security professional I'm not inclined to trust your judgment in security matters any more than a bank president would trust a convicted embezzler. Sorry if that sucks for you, but that's life.
To call me a criminal is rather ignorant and insulting,...
Actually, it's completely accurate. You committed a crime therefore you are a criminal. Stop me if that's too complicated for you to follow.
... I could probably go out and find articles on this or official documentation.
Which is what I asked for, but you have provided a lot of fluff and allegations, but no proof.
...The same law applies to a largely distributed program.
No it doesn't. Sezeniquote...
The Law of Large Numbers: In repeated, independent trials with the same probability p of success in each trial, the chance that the percentage of successes differs from the probability p by more than a fixed positive amount, e > 0, converges to zero as the number of trials n goes to infinity, for every positive e. (bold is mine)
You have made the assumption that the probability of the success of an exploit is constant between IE and Mozilla. That's a bad assumption, and that's where your confusion comes from.
Mozilla code, for reasons listed earlier in this thread, will have a lower exploit rate, thus a lower number of exploits as the number of installations increases.
As the number of installations approaches the number of installations of IE (and I must point out that since IE in integrated into the Windows OS, the number of IE installations will not decrease until the number of Windows installations begins to decrease.) the relative number of exploits will be lower by an increasing factor.
There are many factors which make IE and Mozilla different, including (but not limited to) quality and age of code, complexity, permissions in the OS, speed of patching, number of bug fixers, and so forth.
I still await any documentation that says otherwise.