Path: csiph.com!usenet.pasdenom.info!news.etla.org!news.stack.nl!newsfeed.xs4all.nl!newsfeed1a.news.xs4all.nl!xs4all!post.news.xs4all.nl!not-for-mail
MIME-Version: 1.0
In-Reply-To: <20140701220545.19bf95c9@x34f>
References: <CANc-5UwWVRyQeN15pWSJbeGimPPF7YmBdp6mP-3c5GW7FyK8dQ@mail.gmail.com> <mailman.11380.1404237600.18130.python-list@python.org> <53b30097$0$29985$c3e8da3$5496439d@news.astraweb.com> <20140701220545.19bf95c9@x34f>
From: Ian Kelly <ian.g.kelly@gmail.com>
Date: Tue, 1 Jul 2014 16:34:39 -0600
Subject: Re: What's the "right" way to abandon an open source package?
To: Python <python-list@python.org>
Content-Type: text/plain; charset=UTF-8
Precedence: list
Newsgroups: comp.lang.python
Message-ID: <mailman.11387.1404254127.18130.python-list@python.org>
Lines: 37
NNTP-Posting-Host: 2001:888:2000:d::a6
Xref: csiph.com comp.lang.python:73812

On Tue, Jul 1, 2014 at 1:05 PM, Paul Sokolovsky <pmiscml@gmail.com> wrote:
> On 01 Jul 2014 18:40:23 GMT
> Steven D'Aprano <steve+comp.lang.python@pearwood.info> wrote:
>> http://nedbatchelder.com/blog/201405/github_monoculture.html
>
> Everyone who (re)posts stuff like that should have mandatory N.B. of "I
> just bought a server farm to offer an alternative".

There already are alternatives.

> Second thing is (young pythonistas, close your eyes and ears!) is
> Roooooby. That's where that slight smell of decay which bothers you
> comes from. Indeed, using software written in Ruby is disgusting. But
> if you look around, all software written for advanced git project
> hosting is written in Ruby - gitorious, gitlab. Steven, I hope you have
> plan how to resolve that situation with Python enlightment too. Until
> then people will use that ruby stuff, in particular github.

Ruby has nothing to do with it. (And no, I don't think that using
software written in Ruby is "disgusting".) We're talking about web
sites; from the user perspective, they're just neat bundles of
networking protocols and the implementation language is irrelevant, as
long as their needs are met.

> (Ah, and if
> your solution is "hg", sorry, that won't work - for many reasons,
> including CPython big startup delay :-F.

The warm startup delay for CPython on my box is about 17 ms. hg
starts, prints its list of commands, and exits in about 44 ms. That's
practically invisible to the user, unless you're using a script that
involves executing hg from the shell hundreds of times, in which case
you might want to consider having your script use the command server
instead.

Or if it's the start-up delay on the web server that you're concerned
about, then set up the server to use persistent processes.