Analysis of the reasons why Python3 can revive Python

I read about "Python 3 is destroying Python" from Stephen A. Goss. This article has a lot of wonderful arguments, but I don't think Python 3 is destroying python, and I don't think the whole situation is bad for python at all.

But as the tacky adage says, every crisis may also mean an opportunity.

Perhaps Python 3 can revive Python.

Obviously, the trouble is not just the porting of Python 2 to Python 3. The time is no longer 2005 years, and the young programmer is no longer so excited about which version of Python it is. Yes, there are a lot of Python jobs on the market now, but at the same time there are more Java jobs. And there's a lot of Perl jobs on the market before--it's said to be a bad time. My focus is not on the number of jobs or the number of warehouses on GitHub, my focus is on thought and passion. I know this is a bit subjective, but I feel that Python has recently been missing in both areas.

As an example, we see people moving from Python to go.

Although not many, this phenomenon is noteworthy (including the entire new development team blogging about porting their code base), enough to generate some public opinion (and Rob Pike, who initially wanted people to move from C + + to go).

Python faces challenges from the west. There are challenges in a field that nibble on Python's share (for example, new, out-of-the-box projects will prefer node or go rather than twisted,rails still dominate the web framework), a professional direct competition (for example, Julia of scientific calculation), and the general competition (Clojure,groovy,javascript,dart, etc.).

So here's my idea of Python 3, a very simple idea:

Let Python become Tempting

Python 3 is already incompatible with Python 2, and people are not migrating to Python 3 in droves, so adding some incompatible changes will not only do no harm, but rather benefit the language.

To be honest, if Python 3 has enough attractive new features, more people will be willing to migrate to Python 3 (at least their new project will), and more and more people will be porting their Python 2 libraries and projects. What's more, in the next humble opinion, this will attract more people who are not using Python now.

As you've seen, Python 3 is a tedious update.

True, Python 3 makes Python clearer and fixes some long-standing problems and headaches. But it didn't make that much difference. In the words of Python 3, the idea was set to be relaxed. JavaScript was not so popular at the time. YouTube hasn't been born yet. That was a long time ago.

Now like the right envelope, immutability, good asynchrony, and so on, but the focus of sharp hackers to consider.

Without saying a few words, here are some suggestions that might make Python 3 interesting. At least for me it would interest me:

1. Removes the global interpreter lock (Gil,global interpreter lock). or provide a good asynchronous processing mechanism. Guido Pep 3156 is not sure to solve this problem. It is also good to have a primitive that resembles a channel (channels) like go.
2. Make Python faster. If JavaScript can get faster, CPython can become fast too. Or the pypy can be mature enough to replace the CPython (there should only be one presence). If you need a big coffee like Lars Bak, go kickstarter--I'll pay for it. Let big companies also embarrassed to contribute a little. Not that Dropbox also spends money on developing their Python-based LLVM?
3. Increase the type. Well, it's the opt-in type. This allows you to speed up your code (for CPython), or make sure to help type checking (for dart). Add type annotations to the standard library.
4. Improve the standard library. Find a team to go through these standard libraries, fix long-standing annoying problems, increase speed and fix bugs. Improve the API for these libraries and provide a simpler interface for common things (compare requests and urllib) with the existing standard library to provide a new, improved standard library with a different name domain. And, let their conversions be easy to do (perhaps through some sort of automated tool).
5. Improved "read-evaluate-output loop" (Repl,read-eval-print loop). Please, it's been 2014 years now. Do it again in a modern way repl. Add a little color or something. Please refer to the Ipython. Make it a client/server, so the IDE and the editor can embed it.

So the Python 3 developer, take it slow. Yes, but not too slow. It's about 3 or 4 years. We wait for the ES6 10 years, 3, 4 years we can still wait.

It doesn't look like everyone is using Python 3, so take a little risk. Break. Release it earlier and more frequently. Interact with the community more.

Everyone, Python 3 is actually not really killing python. But it may be able to save it from the things that are killing it.

PostScript: What do you think? Can you think of more tempting ideas for Python 3? What new features will inspire your interest?

