Recs.
Updated
SpecsUpdate
Pros
Cons
Con Speed can matter
Python, like many interpreted and untyped languages, is fairly slow compared to compiled languages like C, C++ or Java. Sometimes this won't matter, but when it does you're talking about twenty times slower than C/C++ and ten times slowed than Java. You need to have a plan for what you will do if it is too slow.
Con Hard to debug other people's code
As the structure of Python code is based on conventions many developers are not following them and so it is difficult to follow/extract the design of not trivial application from the code. While this is a con, I see it in other languages as well. It seems to depend on the programmer. Most people don't learn conventions first, they just start programming. Unless you work for someone who insists you follow the conventions, you will probably go with what you like. You might never look at the conventions.
Con Language fragmentation
A large subset of the Python community still uses / relies upon Python 2, which is considered a legacy implementation by the Python authors. Some libraries still have varying degrees of support depending on which version of Python you use. There are syntactical differences between the versions.
Con Significant whitespace
While proper formatting is essential for any programmer, beginners often have trouble understanding the need and lack the discipline to do it. Add to that all those editors that randomly convert N spaces (usually 8) to tabs and you get an instant disaster. You may need to find yourself an editor/IDE you like and carry it with you on a thumb drive, which isn't a bad idea anyway.
Con Moving large blocks of code in whitespace sensitive languages is scary
Quoting inventor of the V language: "V's syntax is cleaner with fewer rules. Lack of significant whitespace improves readability and maintainability of large code bases and makes generating code much easier. From my experience of working with a huge Python code base, moving large blocks of code in whitespace sensitive languages is scary."
Con Unflexible userbase
You will be expected to rigidly stick to the coding practices and to do everything by-the-numbers. One of the most common complaints I heard from people who left Plone, which is Python based, to Drupal, which is PHP based, is the community is more like a frat house than a community. If you look for help, make sure you follow the rules of whatever type of group you are requesting help from.
Con Too opinionated for a general-purpose programming language
While it's a good language to learn and use after you have mastered a couple of other less rigid programming languages, it's definitely not good for first-time learners. Both the language itself and its community have made it quite clear that you should do everything the "Pythonic way" to get the best results, that it feels more like an opinionated framework instead of a general-purpose programming language, which means if you are a first-time learner and getting too "tuned" to the "Pythonic way" it will be much harder for you to learn other less-opinionated languages compared to the other way around. Like any programming languages and/or frameworks, I'd recommend first-time learners to learn less opinionated ones first to open up your mind, then learn some of the more opinionated ones to increase productivity for specific fields of works.
After all, programming languages are just some utilities for the human mind to interface with the computers, and there are more suitable tools for different tasks, and you should master the "Pythonic way" (after you already have adequate experience in computer programming) instead of locking your mind too close to the "Pythonic way" as a first-time learner.
Con Might not be very future-proof
Lots of features that will probably be crucial as time goes (good support for parallelism for example) are missing or are not that well-supported in Python. Since 2.x and 3.x still exist, be prepared to switch if something makes 3.x take off in the future.
Con Limited support for functional programming
While Python imports some very useful and elegant bits and pieces from FP (such as list comprehensions, higher-order functions such as map and filter), the language's support for FP falls short of the expectations raised by included features. For example, no tail call optimisation or proper lambdas. Referential transparency can be destroyed in unexpected ways even when it seems to be guaranteed. Function composition is not built into the core language. Etc.
Con Does not teach you about data types
Since Python is a dynamically typed language, you don't have to learn about data types if you start using Python as your first language. Data types being one of the most important concepts in programming. This also will cause trouble in the long run when you will have to (inevitably) learn and work with a statically typed language because you will be forced to learn the type system from scratch.
Con Inelegant and messy language design
The first impression given by well-chosen Python sample code is quite attractive. However, very soon a lack of unifying philosophy / theory behind the language starts to show more and more. This includes issues with OOP such as lack of consistency in the use of object methods vs. functions (e.g., is it x.sort() or sorted(x), or both for lists?), made worse by too many functions in global name space. Method names via mangling and the init(self) look and feel like features just bolted on an existing simpler language.
Con Multi-threading can introduce unwanted complexity
Although the principals of multi-threading in Python are good, the simplicity can be deceptive and multi-threaded applications are not always easy to create when multiple additional factors are accounted for. Multi-thread processes have to be explicitly created manually.
Con The process of shipping/distributing software is reatively complicated
Once you have you program the process of having a way to send it to others to use is fragile and fragmented. Python is still looking for the right solution for this with still differences in opinion. These differences are a huge counter to Python's mantra of "There should be one-- and preferably only one --obvious way to do it."
Recommendations
Comments
Flagged Pros + Cons
