Hacker News Reader

Comment by zzzcpan | Hacker News Reader

http://www.simplyhired.com/search?q=perl Showing 1-10 of 220,528 Perl jobs

And you know, Perl is dead.

Basically, Python gave a choice to a lot of people to either invest time into 3.x without any important improvements or to keep using 2.7 and invest time playing with other languages. Which in turn accelerated Python's decline and raised questions about it dying, promoting the idea even more. But there is no need to worry, it's not a bad thing.

View on HN

... python 3 is a really a fairly small effort to learn when you know python 2. You'll learn most of what you need to know in a few days, tops. This comment feels extremely disingenuous

I used to write a lot of python, and the hard part is that as you write code you're pulling in more and more external libraries, and sooner or later one of them will be incompatible with python 3. Then if you want to continue using python 3 in your own code, you need to spend the effort to port that library to python 3.

You don't invest time to learn it, you invest time by writing or converting code.

Writing code for Python 3 and 2 isn't that difficult.

I've been doing it for a while now while working on a variety of different libraries for the Pyramid project, as well as hacking on WebOb which is Python 2 and 3.

Especially if you drop Python 3.2 and lower from the Py 3.x series you have very little trouble writing code that runs on both (with a shim or two here and there).

It's not nearly as difficult as it once was.

https://docs.python.org/2/library/2to3.html

Your snark was somewhat incoherent. Can you be more clear? guessing from the cpan in your name you're a Perl person?

It was coherent and I would not define it as snark. You don't comprehend his point yet you have the wherewithal to label it as incoherent & snarky -- that's snarky. But I digress..

Python users had a choice to stay in 2.7 land or make changes to their codebase for not much in return. Many chose to stay in 2.7 land, which you could safely say is stagnant. As a result, if you can't keep moving with Python, where will you go?

Since Python users have a choice to stick in 2.7 and get no rewards for it or jump into new languages that are evolving, they're going to jump into new languages and these same people are going to start saying things like "python is dead" because in their 2.7 eyes, it is dead.

Finally, unless I'm missing out on some serious deadpan sarcasm, he concludes "Don't worry it's not a bad thing" because look at Perl who everyone says is dead.

> if you can't keep moving with Python, where will you go?

I think it's too soon to tell about the python 2 to 3 switch. I see people on here all the time making remarks about how many years its been and people still aren't on Python 3, but honestly, it's a big change and it takes a while to upgrade.

Python 2 is still supported (till 2020) and that's the line in the sand for a lot of companies. Many companies will have this in their roadmaps, but don't see a need to prioritise upgrading over other business priorities just yet.

I work at a small company for which this is the case. We absolutely will upgrade to Python 3 (before 2020), and can't wait to do so - there's loads of cool stuff in Python 3! But with a large codebase written in 2.7, this is not something we can just do in a weekend sprint. It will take several months of development and careful testing of our many dependencies. I'm sure many companies are in the same boat.

This clearly doesn't mean Python is dying. Both versions of python are still supported, and many, many people are actively using them and producing quality software. It's a great ecosystem to be part of :)

Maybe you think you "comprehend" his point because you decide to interpret it in your own way. But it was not coherent (and honestly I'm not even sure if that was a snark in there or not, mostly because it was incoherent).

No, it was interpreted pretty much correctly, except for the Perl part, which I should clarify.

Perl is definitely dead, I think most people know that. The link was posted to illustrate that other links from that site were painting incorrect picture.

How is a tool like Perl dead when it's still heavily used on hundreds of thousands of machines all over the world? Is C dead as well? I would say, a tool is dead when it's no longer in wide use. Perl is hardly in that category quite yet.

It's dead because new projects aren't started in Perl unlike C.

Your point stated more generally is that while C is the best tool for a wide set of use cases, for each of the one-time use cases of Perl there is now a better option

Are you saying it's a better investment to jump to another language altogether than to jump from 2.7 to 3.5?

It wasn't incoherent to me - he used Perl to argue that just because a language has a high number of open jobs on simply hired doesn't mean that the language is not dead.

> Showing 1-10 of 220,528 Perl jobs

Not really. On page 18 you already have jobs like http://www.simplyhired.com/job/net-developer-job/compugain-l... were Perl is just used as an example for scripting languages.

The exact same could be said about Perl 6 ;)

It can't be said - Perl6 is completely different language and brings a lot of desireable features on the table. The problem is that it is not so easy to port Perl5 code to Perl6 as porting Python from 2 and 3.

> The problem is that it is not so easy to port Perl5 code to Perl6 as porting Python from 2 and 3.

Is this really the case?

https://github.com/niner/Inline-Perl5

http://stackoverflow.com/questions/9173043/how-can-i-use-per...

Yes this is the case; Through libraries we are trying to make it easy to CALL perl5 (and ruby and python and...) modules from Perl 6. But porting -- aka translating -- is another matter. Mostly it requires a similar amount of effort as if you are translating your program to any other dynamic language for nontrivial programs.

My understanding is that LOTS of Python2 programs work fine in Python3 with no (or very trivial) changes. Very very few Perl5 programs will work fine in Perl6 with no changes.

While this is true, a good deal of porting could be managed by packaging up old code into Perl5 modules and using Perl6 to glue it together, no?

Reply on news.ycombinator.com