Although we've answered Eric's original question to over the hills and far
away and all the way back, I feel compelled to point out that python 2.7's
end-of-life support is on the horizon:

https://pythonclock.org/

While existing python 2.x installations won't turn into pumpkins in 2020,
you won't be getting the benefit of updates.

cheers,

AC

On Wed, Mar 7, 2018 at 3:35 PM, Bryan Brown <[log in to unmask]> wrote:

> Hi Eric,
>
>
> Unless you are specifically working on a legacy project written in Python
> 2 that is to large to update, I would strongly recommend working with 3 as
> its the future. Staying with 2 means you won't have access to cool new
> features in 3.
>
>
> I ran into issues when I started working in Python 3 as it would sometimes
> conflict with whatever the default system version of Python was, whether
> that be 2 or even an older version of 3 than what I prefer. My solution has
> been to work with a tool called "Anaconda", which is a self-contained
> Python distribution that you can download into your user account instead of
> the system itself, which avoids conflicts with the system Python. Anaconda
> also has an extended package index, so you can quickly install data
> sciencey modules much easier than with default Python. I highly recommend
> checking it out.
>
>
> https://www.anaconda.com/download/
>
>
> Bryan J. Brown
>
> Repository Developer
>
> Technology & Digital Scholarship Division
>
> Florida State University Libraries
>
> ________________________________
> From: Code for Libraries <[log in to unmask]> on behalf of Eric
> Lease Morgan <[log in to unmask]>
> Sent: Wednesday, March 7, 2018 3:14:23 PM
> To: [log in to unmask]
> Subject: [CODE4LIB] python 2 versus python 3
>
> I need some advice regarding writing in Python 2 versus Python 3.
>
> I have been trying to write more of my software in Python because
> everybody around me is speaking French (Ruby), Italian (R), or Spanish
> (Javascript) when I’m speaking in Latin (Perl). My latest example is a set
> of command-line tools (Bash as well as Python) designed to do scalable
> reading against HathiTrust content. [1] I originally wrote the tools in
> Python 2. Since then I learned more about the HathiTrust Research Center
> “capsules” (on-demand virtual machines), and there Python 3 is the default
> environment. So, I tweaked my code and it works out-of-the-box and as
> expected.
>
> On the other hand, my kewl tools no longer work in my local environment
> because I use Python 2, and I don’t feel like installing YAVoP (Yet Another
> Version of Python) on my server(s).
>
> What’d do you suggest? Do you suggest writing in: 1) Python 2 or 2) Python
> 3 or 3) a combination of both?
>
> [1] HathiTrust Research Center Workset Browser -
> https://urldefense.proofpoint.com/v2/url?u=https-3A__github.
> com_ndlib_text-2Danalysis-2Dhtrc&d=DwIFaQ&c=HPMtquzZjKY31rtkyGRFnQ&r=
> FkFKSediiYRWi53Dc5NSDQ&m=zN3GlgstLlVZyTx8TpTmciwPYI1PKKXPdAnYA5KVZlY&s=
> oGhXqGrquStoMR-DYzbkkVBGqGe-em7LB3PAtVn9OEE&e=
> [2] HathiTrust Research Center “capsules” - https://urldefense.proofpoint.
> com/v2/url?u=https-3A__analytics.hathitrust.org_staticcapsules&d=DwIFaQ&c=
> HPMtquzZjKY31rtkyGRFnQ&r=FkFKSediiYRWi53Dc5NSDQ&m=
> zN3GlgstLlVZyTx8TpTmciwPYI1PKKXPdAnYA5KVZlY&s=z3Y98jYwrJBscwfAvzxsz7-
> i4iiwalm2l2uq5EW2UNA&e=
>
> —
> Eric Morgan
>