4 ms·

That doesn’t match my recollection. I’m not sure how to prove anything, but I’ll note that Python 3.0 was released in 2008. If Python 2 wasn’t already extremely

by wk_end 3y ago

That doesn’t match my recollection. I’m not sure how to prove anything, but I’ll note that Python 3.0 was released in 2008. If Python 2 wasn’t already extremely popular and well-established at that point, I doubt we’d have seen the brutal 10+ish year migration from it.

littlestymaar 3y ago
> I’m not sure how to prove anything, but I’ll note that Python 3.0 was released in 2008. If Python 2 wasn’t already extremely popular and well-established at that point, I doubt we’d have seen the brutal 10+ish year migration from it. This is about the “scripting niche” I was talking about: in 2008 every Linux distro included Python and lots of build scripts were written in Python 2, but Python for the back-end was really rare back then.
- scrollaway 3y ago
  Yeah python for the backend started taking off with Django, which only released 1.0 and picked up steam in 2008.
  - qup 3y ago
    Django was already very good well before 1.0, and had at a minimum some of the best documentation out there. I seem to recall a decent community, video tutorials, etc.
nostrademons 3y ago
It depends how you define "mainstream". Python was definitely around, had a very strong community, and a number of very significant programs written in it. (Among them: the first versions of BitTorrent and Google.) However, if you were a random dev applying for a random job, you would most likely have to work in C++, Java, or C#. It was relatively difficult getting a job where you would write the majority of your code in Python. Certainly they existed, and many jobs would have you write small one-off scripts in Python, but the actual core product would usually be one of Java/C#/C++. Remember that in 2009, Google had 20K employees while IBM (which was a major booster of Java) had 400K. And at Google, Python wasn't really allowed for major projects (Search had been rewritten from Python to C++ when it got big in 2000), so you'd be using Java or C++ anyway. Data science wasn't really a thing, and if you did machine learning you used frameworks like Weka or home-grown stuff.

by wk_end 3y ago

littlestymaar 3y ago

> I’m not sure how to prove anything, but I’ll note that Python 3.0 was released in 2008. If Python 2 wasn’t already extremely popular and well-established at that point, I doubt we’d have seen the brutal 10+ish year migration from it. This is about the “scripting niche” I was talking about: in 2008 every Linux distro included Python and lots of build scripts were written in Python 2, but Python for the back-end was really rare back then.

scrollaway 3y ago
Yeah python for the backend started taking off with Django, which only released 1.0 and picked up steam in 2008.
- qup 3y ago
  Django was already very good well before 1.0, and had at a minimum some of the best documentation out there. I seem to recall a decent community, video tutorials, etc.

nostrademons 3y ago

It depends how you define "mainstream". Python was definitely around, had a very strong community, and a number of very significant programs written in it. (Among them: the first versions of BitTorrent and Google.) However, if you were a random dev applying for a random job, you would most likely have to work in C++, Java, or C#. It was relatively difficult getting a job where you would write the majority of your code in Python. Certainly they existed, and many jobs would have you write small one-off scripts in Python, but the actual core product would usually be one of Java/C#/C++. Remember that in 2009, Google had 20K employees while IBM (which was a major booster of Java) had 400K. And at Google, Python wasn't really allowed for major projects (Search had been rewritten from Python to C++ when it got big in 2000), so you'd be using Java or C++ anyway. Data science wasn't really a thing, and if you did machine learning you used frameworks like Weka or home-grown stuff.