It became clear that to build a successful mobile application you should make it available on all of the major mobile marketplaces - namely Apple AppStore, Google PlayStore and Microsoft Windows Store.
Err... major mobile marketplaces and Windows Store in one sentence? Sounds like bullshit to me. Android users are leechers and do not pay for apps and ios app coders do not need a multi device framework because they have a fairly homogenous environment. Also only about 5 different screensizes instead of a brazillion different ones for android.
*spills flamewar fuel all around the room*
Heh... well, dunno about others, but I've paid for numerous applications on my Android.
If you break that down to $/download it looks pretty lousy.
That's probably more than I've paid for software for my desktop in that period, honestly.
Yes, ads might be a source of additional income. I never click them.
And I also spent most of my app money on games, but mostly for my son. :(
I got 19 of 20. Also missed a real Basic variety.
I have a fun little math problem I want to solve for a real-world application (not for where I work right now).
I have a file that could be up to 10 gigs large.
I want to break that file down to x number of chunks of y bytes. I need for y to be divisible by 8, but it's okay if the chunk comes up short (say, if the file doesn't neatly divide by y, or by 8).
The tricky bit is to ensure x and y are as close as possible to each other in value. They don't have to be exact, but that's the ideal.
That's the puzzle I'm working on in my spare time.
Hmm... now I see why that's so hard to do programmatically.
Let x = size of data chunks divided by 8.
Let y = number of data chunks.
Let t = total size of the data.
(x / 8) * y = (t/8)
(x * (8 * y) = t
x = t / (8 * y)
or, viewed another way:
x = (t / 8) * y
So, after dividing t by 8, you take that number, break it down to its various primes, then figure out how to divvy up the two sets of primes such that if you add the primes together, they're as close as possible to the same result.
An example: if t == 100, then x = (100 / 8) * y, or x = 12 / y. If x = 4 and y = 3, then 4 = 12 / 3, a valid answer.
When applied to the problem above, 4 * 8 = 32, 3 * 32 = 96, which is pretty close to 100.
So, 4 chunks of 32 bytes will cover the 100 byte size, breaking 100 up nicely, while not using too many chunks.
The hard part is going to be that breaking down the primes and finding a combination of them that comes up close to the same number. I seem to recall calculating the primes for a particular number isn't particularly simple in the first place. Still, that's the optimum thing that I need. Hrm.
Yeah, looks like prime factorization is part of what I need to do this efficiently, and greater minds than mine have an interest in solving that efficiently for large numbers.
A 64-bit number might not be considered 'large' for this, though. But the algorithms I've seen for solving this (currently) seem arcane. As in, if I could fully grok those algorithms, I should change careers.
Not that most people here would care, but C++14, as of yesterday, is now a standard that will be published later this year:
This is kind of amazing, in that this standard is being published at about the same time that the people who make compilers are releasing versions of the compiler that conform to it. In the past, you had to wait years before you had a compiler that matched the standard.
As for what is in C++14, I don't know yet. I haven't been keeping close enough track to see what they're doing. I suspect they have been trying to focus on multi-processor related commands (atomic memory management, a clearer handling for threads and sychronization, etc).
ipython notebook seems to be cute.
I'm not current on the religious wars. Does python count as programming?
Also, ipython is really amusing.
Python is used quite a bit in the industry... no controversy there.
Although I think there's a bit of a scuffle between Python and Perl, as they are both popular scripting languages.
well, i'd rather say, there used to be controversy.
perl has lost.
And that for a reason. while its easy to write perl programs that look to anybody else like character soup, its easy in python to write human(!) readable programs.
Python is definitely a good language to start learning to program.