Anthony Scopatz

I think, therefore I amino acid.

Python 3000

So I have been reading about Python 3000, the forthcoming next stage of my favorite programming language. Initially I was appalled, frightened. They are doing a couple of things that are scary but take a lot of courage. First, they are breaking backwards compatibility. Furthermore, knowing how I code, nothing I have written so far is inane enough to make it past this barrier. This partially because they are making print into a function rather than a syntactical element like for, if, etc. This really annoyed me until (who wants to type parentheses anyways?) they pointed out that as a function it could be overwritten in your module! How fun would that be to have a version of print that highlights every letter ‘e’ in red when it prints. Furthermore, I think this helps with some of the otherwise weird issues in running external programs inside of python with respect to printing stout.

However, one of the main philosophies about Python is that everything (that can be) is a list. This makes refereeing to elements inside of otherwise bulky structures ease, and iterating over them simple (imported files for example are lists of lines read from that file, or can be). This adherence is going away and being replaced for some structures (like dictionarys) with iterators. The dict.keys() object won’t be a list…which is odd, but instead will be an iterator over the list of the dictionary’s keys. Similarly, range() will be an iterator that is know known as xrange(). Iterators tend to be more effiecent than lists, especially for large lists. This is good as it will force me, and others, to write better code. I like the idea of going to iterators where you can. I hope they make this a full Philisophical change though, rather than simple a couple, often used special cases.

Python is amazing because of its solidness of philosophy and adherence not “natural coding”. The obvious fear is that if the excitement to come up with something better they may make the product more unusable. All-in-all I am still excited to see how it all works. It should be fun to dick around with at the least. It just seems that all of my favorite CS projects out there are changing or dying. First my window manager has ended support and the project is dead. Python is changing. Gentoo had some internal shake up earlier this year. If all I have to look forward to is Ubuntu on a Dell, then I will be one Peeved Panda v16.