A company is about 3 things:
- Delivering value
- Delivering quality (holding value over time)
- Delivering within constraints (delivery time/money/scope)
Python makes it possible to supply value with sufficient quality within restrictions, as long as the delivery time is not under pressure.
In many cases, the speed is subordinate to the value creation.
In addition, speed can be achieved in multiple ways.You can have 1 CPU run 20 calculations, or 20 Cpus 1 calculation. In The latter case, an individual task may take up to 20 times longer, and still the same turnaround time is possible.
As long as you can make a solution that uses a lot of loose computers/computing power, the speed limitation is relative.
However, there is another aspect to Python.
It is popular because it is a reasonably simple programming language.
You can learn the basics with a few hours if you have some programming experience.And with that you can get started, especially with the documentation and support that is (free) available on the Internet.
This enables the very high demand for software developers, the learning of other (faster) programming languages (Java/C/C++) often takes much more time.
This is one aspect we shouldn’t underestimate, something similar happened in the late years 90 and early 00 鈥瞫 in the General IT. There was just not enough staff to find an IT training to meet the demand.And so, people from other disciplines were retrained into IT staff. Then you need simple solutions that are fast to learn, no complex solutions that are superior, but also cost three times as much time to learn.
Because the speed of a programming language really matters only in a limited number of niches.
In Most cases there are things like readability, availability of good ‘ packages ‘, etc.Much more important, and Python scores particularly well on those planes.
It’s at run fast enough.
And development at lightning speed.There are many libraries available and the language itself is very powerful. I did read that a Java programmer needs 3x as many rules as a Python programmer with 3x the chances of bugs. Is fixed too short by the bend but it does give an idea….
M.I. Python is a relatively simple “entry level” that you can accomplish a lot of thanks to countless already-used tools and libraries.Especially if you know what you are doing, and consider disadvantages such as relatively small call stack or the existence of supposedly Global Interpreter Lock, or GIL, which can make multithreading challenging.
Possible use cases:
- Data crunching.
Most ML packages have a Python interface.
For example, Flask (a Web framework in Python) can quickly build your MVP.
Let’s face it, Python is already more popular than awk or Perl, and there are tools, like (for example) Ansible, with which you can easily get started as devops.
(NB: Apologies for any anglicisms, Dutch my mother tongue is not:))