Quote:
Originally Posted by Katsunami
The most notable is that C++ requires you to handle a lot of stuff yourself, which Python handles "on the fly" while executing the program. This makes C++ much harder to write a program in because you need to do (almost) everything, down to the smallest detail, but if you get it done right, it will be much faster than the Python equivalent.
Therefore, if you need the absolutely fastest program you can get (think about a chess engine, for example, calculating thousands of moves per second), you choose C++. If you need to be able to write a program without bothering with the nitty-gritty tiny details, and speed is not of paramount importance, you go for something such as Python.
If Calibre was written in C++, it would have been faster. Also, you would probably see an update once every month instead of once every week, and there would probably be MUCH less plugins available.
|
They are both using Qt though and in my experience there is no significant overhead using C++/Qt over PyQt. It depends on what you are more familiar with. Btw, having the main application in C++ does not mean that plugins have to be also in C++, although I don't know if it will be possible to maintain compatibility with Calibre plugins.
Quote:
|
Originally Posted by Katsunami
He aspires to replace Calibre with his own product.
|
They do mention on their website that they are using calibre for some tasks and give kudos to Kovid for it. They don't even describe it as an alternative to Calibre but as a compliment[sic] to it. Even if they aspire to replace Calibre at some point, what is the problem with that? It might work better for some people and it makes sense to slowly replace parts instead of rewriting everything in one go.
Anyway I don't understand all the negativity. BookONO is in early alpha and people are already dooming it to failure instead of encouraging the developers. Calibre is great and all but it is not beyond improvement - otherwise you wouldn't be upgrading it every so often. Not to mention that it is virtually impossible to satisfy every user 100%. Any alternative that comes out is a net positive in my view.