Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #87339 > unrolled thread

Google Code Shutting Down

Back to article view | Back to comp.lang.python

Contents

  Google Code Shutting Down Josh English <Joshua.R.English@gmail.com> - 2015-03-12 15:26 -0700
    Re: Google Code Shutting Down Mario Figueiredo <marfig@gmail.com> - 2015-03-13 00:12 +0100
      Code hosting providers (was: Google Code Shutting Down) Ben Finney <ben+python@benfinney.id.au> - 2015-03-13 12:37 +1100
        Re: Code hosting providers (was: Google Code Shutting Down) Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2015-03-13 14:17 +1100
          Re: Code hosting providers Michael Torrie <torriem@gmail.com> - 2015-03-12 21:28 -0600
          Re: Code hosting providers Ben Finney <ben+python@benfinney.id.au> - 2015-03-13 15:22 +1100
          Re: Code hosting providers Ben Finney <ben+python@benfinney.id.au> - 2015-03-13 15:27 +1100
        Re: Code hosting providers Paul Rubin <no.email@nospam.invalid> - 2015-03-12 22:17 -0700
          Re: Code hosting providers Chris Angelico <rosuav@gmail.com> - 2015-03-13 19:33 +1100
            Re: Code hosting providers Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2015-03-13 19:59 +1100
              Re: Code hosting providers Chris Angelico <rosuav@gmail.com> - 2015-03-13 22:46 +1100
                Re: Code hosting providers Paul Rubin <no.email@nospam.invalid> - 2015-03-13 14:13 -0700
                  Re: Code hosting providers Gene Heskett <gheskett@wdtv.com> - 2015-03-13 19:38 -0400
                    Re: Code hosting providers Mario Figueiredo <marfig@gmail.com> - 2015-03-14 01:48 +0100
                      Re: Code hosting providers Paul Rubin <no.email@nospam.invalid> - 2015-03-13 19:30 -0700
                      Re: Code hosting providers Gene Heskett <gheskett@wdtv.com> - 2015-03-13 23:13 -0400
                        Re: Code hosting providers Mario Figueiredo <marfig@gmail.com> - 2015-03-14 04:41 +0100
                          Re: Code hosting providers Paul Rubin <no.email@nospam.invalid> - 2015-03-13 20:45 -0700
                            Re: Code hosting providers Rustom Mody <rustompmody@gmail.com> - 2015-03-13 22:03 -0700
        Re: Code hosting providers (was: Google Code Shutting Down) Mario Figueiredo <marfig@gmail.com> - 2015-03-13 11:02 +0100
        Re: Code hosting providers Paul Rubin <no.email@nospam.invalid> - 2015-03-13 14:35 -0700
    Re: Google Code Shutting Down Ian Kelly <ian.g.kelly@gmail.com> - 2015-03-12 22:33 -0600
      Re: Google Code Shutting Down Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2015-03-13 19:47 +1100
    Re: Google Code Shutting Down Josh English <Joshua.R.English@gmail.com> - 2015-03-13 12:53 -0700

Page 1 of 2  [1] 2  Next page →

#87339 — Google Code Shutting Down

I've been hosting Python projects on Google Code, and they're shutting down.

Damn.

What is the recommended replacement for Code Hosting that works reliably with PyPi and pip?

[toc] | [next] | [standalone]

#87341

On Thu, 12 Mar 2015 15:26:08 -0700 (PDT), Josh English
<Joshua.R.English@gmail.com> wrote:

>I've been hosting Python projects on Google Code, and they're shutting down.
>
>Damn.
>
>What is the recommended replacement for Code Hosting that works reliably with PyPi and pip?

Essentially anywhere where either Git, Bazaar, Mercurial or Subversion
are supported. See:

https://pip.pypa.io/en/latest/reference/pip_install.html#vcs-support

[toc] | [prev] | [next] | [standalone]

#87344 — Code hosting providers (was: Google Code Shutting Down)

Mario Figueiredo <marfig@gmail.com> writes:

> On Thu, 12 Mar 2015 15:26:08 -0700 (PDT), Josh English
> <Joshua.R.English@gmail.com> wrote:
>
> > What is the recommended replacement for Code Hosting that works
> > reliably with PyPi and pip?
>
> Essentially anywhere where either Git, Bazaar, Mercurial or Subversion
> are supported.

Installing directly from whatever the latest revision in VCS, is not a
good replacement. What's needed is a VCS hosting provider that also
expects you to publish tarballs at specific versions.


For that, you want to choose a provider that runs entirely free software
for the service – not just the VCS, but whatever other features you use
need to be freely available for someone to run another instance of the
same service.

Otherwise you're stuck any time you want to migrate your project, which
is what led to this thread in the first place.

This is called “vendor lock-in”: services like GitHub and Google Code
lock you in to the one provider, because if you want to migrate you
can't take your project data (not just VCS but also issues, merge
requests, etc.) with you to continue somewhere else.

Any service which doesn't run their service on free software is one to
avoid <URL:http://mako.cc/writing/hill-free_tools.html>; free software
projects need free tools to remain that way.


GitLab <URL:https://about.gitlab.com/> is a good option: they provide
VCS, file hosting, wiki, issue tracker, code review via merge requests,
etc. and all of it can be migrated to any other instance of the same
service.

Also worth watching is Kallithea, a new federated code hosting service
<URL:https://kallithea-scm.org/>. It supports Mercurial and Git for VCS,
code review, and integrates with existing issue trackers. Because it's
federated, you won't suffer from vendor lock-in.

Good hunting in finding a free-software code hosting provider for your
projects!

-- 
 \      “I knew things were changing when my Fraternity Brothers threw |
  `\           a guy out of the house for mocking me because I'm gay.” |
_o__)                                      —postsecret.com, 2010-01-19 |
Ben Finney

[toc] | [prev] | [next] | [standalone]

#87347 — Re: Code hosting providers (was: Google Code Shutting Down)

Ben Finney wrote:

> Also worth watching is Kallithea, a new federated code hosting service
> <URL:https://kallithea-scm.org/>. It supports Mercurial and Git for VCS,
> code review, and integrates with existing issue trackers. Because it's
> federated, you won't suffer from vendor lock-in.


What do you mean by "federated"?


-- 
Steven

[toc] | [prev] | [next] | [standalone]

#87348 — Re: Code hosting providers

On 03/12/2015 09:17 PM, Steven D'Aprano wrote:
> Ben Finney wrote:
>> Also worth watching is Kallithea, a new federated code hosting service
>> <URL:https://kallithea-scm.org/>. It supports Mercurial and Git for VCS,
>> code review, and integrates with existing issue trackers. Because it's
>> federated, you won't suffer from vendor lock-in.
> 
> What do you mean by "federated"?

I think it's a fancy way of saying you can run your own instance of the
full web system on your own server under your own domain name.

[toc] | [prev] | [next] | [standalone]

#87351 — Re: Code hosting providers

Steven D'Aprano <steve+comp.lang.python@pearwood.info> writes:

> Ben Finney wrote:
>
> > Also worth watching is Kallithea, a new federated code hosting service
> > <URL:https://kallithea-scm.org/>. It supports Mercurial and Git for VCS,
> > code review, and integrates with existing issue trackers. Because it's
> > federated, you won't suffer from vendor lock-in.
>
> What do you mean by "federated"?

A service is federated if an independent, perhaps even competing, vendor
can provide the same service and all users of the service participate,
moving freely from one vendor to another as they choose without losing
any of their data or identity.

Another term is “decentralised”, though that says only what the service
is not. I prefer the term “federated” because it describes what the
service *is*: composed of multiple parties, collaborating on equal terms
enforced for the protection of the people participating in the service.

If a service's features use proprietary protocols or undisclosed APIs or
in any other way make it infeasible for a user joining today to reliably
migrate their accumulated data to a different vendor at some future
time, then that service is not federated and its users are vulnerable to
vendor lock-in.

The email system, for example, is federated: my <ben@benfinney.id.au>
data can reliably (and has, several times) survive migration to any
vendor who provides standard email service, because I lose no access to
any of my history or identity or accumulated data or access or ongoing
functionality with the same data. By “reliably” I mean that I can move
it all from one vendor to another, and even back again if I choose, in
the knowledge that the standard POP, IMAPv4, and SMTP protocols will be
honoured, because to fail in that would be a bug severe enough that the
vendor would be avoided as not providing the service.

Other examples of federated systems include:

* Avatar image serving: Libravatar <URL:https://www.libravatar.org/>.

* Social networking: Diaspora <URL:https://joindiaspora.com/>.

* Version control: Git <URL:http://git-scm.com/>.

* The World Wide Web, and the internet as originally designed
  <URL:http://www.theguardian.com/technology/2012/sep/05/tim-berners-lee-internet-off-switch>.


To the extent that vendor lock-in is possible, the service is not
federated. Corollary: some features of a vendor's offerings can be
federated, and others not; and to talk of the vendor's complete set of
offerings as “federated” can only be true if they all are.

The GitHub service, for example, is not federated, because while mere
Git repositories *are* federated, significant GitHub features (issue
tracker, pull requests, and other important history and project data)
cannot reliably be migrated smoothly to another instance provided by a
competing vendor.

Ned Batchelder and Nick Coghlan have written insightfully on this
<URL:http://nedbatchelder.com/blog/201405/github_monoculture.html>
<URL:https://mail.python.org/pipermail/python-dev/2014-November/137153.html>
with regard to why GitHub – or any non-federated service – is not a good
option for projects that intend to stay viable.


It will be good to see Kallithea <URL:https://kallithea-scm.org/> come
into its own; until then we have GitLab <URL:https://about.gitlab.com/>
and Redmine <URL:https://www.redmine.org/> and even the juggernaut of
Launchpad <URL:https://launchpad.net/>; they are all federated and avoid
vendor lock-in.

We should ask hard questions to projects that choose to host their
irreplaceable data – not just VCS but issues, user identities,
discussions, documentation, etc. – on services that they can't reliably
migrate from. They are demonstrating they don't intend to be around for
long; why should we invest any effort in them?

-- 
 \      “The most common way people give up their power is by thinking |
  `\                               they don't have any.” —Alice Walker |
_o__)                                                                  |
Ben Finney

[toc] | [prev] | [next] | [standalone]

#87352 — Re: Code hosting providers

Michael Torrie <torriem@gmail.com> writes:

> On 03/12/2015 09:17 PM, Steven D'Aprano wrote:
> > What do you mean by "federated"?
>
> I think it's a fancy way of saying you can run your own instance of the
> full web system on your own server under your own domain name.

With the corollary that *any* vendor motivated to do so can serve the
same function. Meaning the options aren't just “use this vendor or get
nothing”, as they are with a centralised service.

With a federated service, anyone *can* run their own instance, which
means that few people have to: vendors can pop up and compete with one
another, participants have the option of moving to a different vendor
taking their data and functionality and established relationships and
discussions and identity where they choose.

-- 
 \     “The aim of science is not to open the door to infinite wisdom, |
  `\        but to set some limit on infinite error.” —Bertolt Brecht, |
_o__)                                        _Leben des Galilei_, 1938 |
Ben Finney

[toc] | [prev] | [next] | [standalone]

#87354 — Re: Code hosting providers

Ben Finney <ben+python@benfinney.id.au> writes:
> Any service which doesn't run their service on free software is one to
> avoid <URL:http://mako.cc/writing/hill-free_tools.html>; free software
> projects need free tools to remain that way.
>
> GitLab <URL:https://about.gitlab.com/> is a good option: they provide
> VCS, file hosting, wiki, issue tracker, code review via merge requests...
> Also worth watching is Kallithea, a new federated code hosting service
> <URL:https://kallithea-scm.org/>. It supports Mercurial and Git for VCS,
> code review, and integrates with existing issue trackers. Because it's
> federated, you won't suffer from vendor lock-in.

Thanks for that informative post: I've been uncomfortable with the reach
of Github and you've done a good job explaining the reasons.

I personally use self-hosted git repositories on cheap VPS servers which
is easy to do, but doesn't supply issue tracking.  I know various
projects that use either Trac or Bugzilla but I haven't tried operating
those.

One interesting and very lightweight alternative is Fossil, which is an
all-in-one program that does lots of the same stuff as Git, Trac, a web
server, and a wiki, all rolled into a single C program that uses about 1
meg of ram.  It's written by the guy who wrote SQLite and it in fact
uses SQLite as the underlying data store.  It has a philosophical
difference with git in that it's difficult to "rewrite history" which
some workflows expect you to do.  I've sometimes wondered how difficult
it would be to modify Fossil to support that feature.

[toc] | [prev] | [next] | [standalone]

#87360 — Re: Code hosting providers

On Fri, Mar 13, 2015 at 4:17 PM, Paul Rubin <no.email@nospam.invalid> wrote:
> Thanks for that informative post: I've been uncomfortable with the reach
> of Github and you've done a good job explaining the reasons.
>
> I personally use self-hosted git repositories on cheap VPS servers which
> is easy to do, but doesn't supply issue tracking.  I know various
> projects that use either Trac or Bugzilla but I haven't tried operating
> those.

It's worth noting, by the way, that sometimes it's worth using a
non-federated service. I host most of my projects on GitHub, because
the git repo is the part that's most important to me. I don't heavily
use the bug tracker attached to any of those projects, nor the
gh-pages branch, etc, etc. If ever I need to move off GitHub, the loss
of my tracker history is almost entirely non-significant. But if
someone goes looking at my repo there, and wants to raise an issue,
I'm quite happy to have that come through the tracker - which means
that, technically, I am using that non-federated service. It's not
something to fear, just something to be aware of.

With the example of moving emails around, the hard part usually isn't
SMTP and co, but moving your email history. Can you download all your
emails in a convenient format for unloading into your new host, or do
you get a proprietary .PST file that you have to somehow deal with?
I'm not speaking from any sort of experience here, of course. It's not
like I've had to deal with several people who decided not to move to
an open system because it would be too hard to move their data. No, of
course not. That never happens.

ChrisA

[toc] | [prev] | [next] | [standalone]

#87363 — Re: Code hosting providers

Chris Angelico wrote:

> It's worth noting, by the way, that sometimes it's worth using a
> non-federated service. I host most of my projects on GitHub, because
> the git repo is the part that's most important to me. I don't heavily
> use the bug tracker attached to any of those projects, nor the
> gh-pages branch, etc, etc.


That is the complete opposite of making sense.

If you had said, "Sometimes it's worth using a non-federated service, and
risking vendor lock-in, because the extra features they provide are just
that good" then I'd accept that. That makes sense. I don't like it, but
that's the business model of proprietary services: provide more features,
and use that to lock people in.

But saying "I host my projects on GitHub because I DON'T use any of the
features which differentiate GitHub from its federated competitors" makes
no sense to me. You are actively helping to support a software monoculture,
and you're not even getting any short-term benefit from it! That's the
worst of all possible worlds -- selling out the future, for no gain today.

Personally, I won't use Github if I can possibly ignore it. I have ethical
issues with their corporate culture, and I dislike anything which smacks of
a monoculture. The more Github gets treated as synonymous with revision
control, the more I will push back, until I cannot push back any more.


-- 
Steven

[toc] | [prev] | [next] | [standalone]

#87368 — Re: Code hosting providers

On Fri, Mar 13, 2015 at 7:59 PM, Steven D'Aprano
<steve+comp.lang.python@pearwood.info> wrote:
> If you had said, "Sometimes it's worth using a non-federated service, and
> risking vendor lock-in, because the extra features they provide are just
> that good" then I'd accept that. That makes sense. I don't like it, but
> that's the business model of proprietary services: provide more features,
> and use that to lock people in.
>
> But saying "I host my projects on GitHub because I DON'T use any of the
> features which differentiate GitHub from its federated competitors" makes
> no sense to me. You are actively helping to support a software monoculture,
> and you're not even getting any short-term benefit from it! That's the
> worst of all possible worlds -- selling out the future, for no gain today.

Not quite. It's a matter of priorities. I will make use of their
proprietary features, but they aren't important to me, and if ever I
need to move away from GitHub, I'll just shrug and abandon all of that
ancillary stuff. In the meantime, I get zero-dollar hosting of my
repos, including zip download and such (I'm not sure how many other
hosts have that, but it's a minor convenience rather than anything
crucial), and it's a convenient place to point people.

You're welcome to shun them. There is definitely benefit to
encouraging a multiplicity of hosting services. But I'm not bothered
by the GitHub non-free-ness, because I take a less philosophical and
more pragmatic view of things.

ChrisA

[toc] | [prev] | [next] | [standalone]

#87384 — Re: Code hosting providers

Chris Angelico <rosuav@gmail.com> writes:
> In the meantime, I get zero-dollar hosting of my repos, including zip
> download and such ...  You're welcome to shun them. There is
> definitely benefit to encouraging a multiplicity of hosting
> services. But I'm not bothered by the GitHub non-free-ness, because I
> take a less philosophical and more pragmatic view of things.

But you're making the problem worse for other people too.  For example,
I can't file a bug against your code or send you a pull request, without
enrolling in Github myself.

You can get a whole virtual server from Amazon for free that is
perfectly good for hosting git repos and lots of other things too, or
you can get them for a few dollars a year in other places if you don't
want to deal with Amazon.  Github adds some conveniences but I've never
understood the attraction and mystique around it.

[toc] | [prev] | [next] | [standalone]

#87396 — Re: Code hosting providers

On Friday 13 March 2015 17:13:41 Paul Rubin wrote:
> Chris Angelico <rosuav@gmail.com> writes:
> > In the meantime, I get zero-dollar hosting of my repos, including
> > zip download and such ...  You're welcome to shun them. There is
> > definitely benefit to encouraging a multiplicity of hosting
> > services. But I'm not bothered by the GitHub non-free-ness, because
> > I take a less philosophical and more pragmatic view of things.
>
> But you're making the problem worse for other people too.  For
> example, I can't file a bug against your code or send you a pull
> request, without enrolling in Github myself.

That alone is grounds enough to get the heck out of github.  This is 
classic vendor lockin, or lockout if you don't want to play by githubs 
rules.  A classic demonstration of TANSTAAFL if there ever was one.

The above quoted problem YOU, Chris do not see, but that should be reason 
enough to bail out AND take your code with you before you find yourself 
locked out of your own code.

Running your own server is a piece of cake, and if I, at 80 yo, can do 
it, I don't see a single reason you can't do likewise.  The code I 
write, for what is called a legacy computer, is just one of the things I 
share at the link in my sig.  That link is actually _this_ machine.  
Sure, I blow my own horn a bit, & you have to look at my ugly mug while 
it loads on the front page, but that is the only advertising you will 
ever see on that site.  If I am still breathing regular when the 5 year 
namecheap registration expires, I'll renew it for another 5 years.  I'll 
be approaching 84 yo then.

That namecheap registration, and the electricity to run it are my only 
extra expenses over and above my use of the machine as a web tourer, and 
an email system.  Or as a remote terminal into one of my milling 
machines so I can write gcode and test execute it here in a comfy office 
chair, which beats the hell out of extended periods standing in front of 
the machines own keyboard.  An hour of that at my age is very painfull.

But don't even think of using the std http port 80, thats blocked at your 
isp, precisely to force you to use, and pay for, space on their servers.

Your /etc/services contains a list of ports, and what they are supposed 
to be used for. The 6309 that I use, is the simplified jedec number of a 
cpu chip that can be transplanted into a TRS-80 Color Computer to 
replace its normal 6809, is 10%  faster at the same clock speed, and 
often 4x faster than a 6809 with some of its added instructions.  It was 
not listed in /etc/services, so there was zero reason not to use it as I 
wasn't stepping on any toes to use it.

Surely this group can look at the ascii code for the word python, and 
come up with a port number that is unique AND not in /etc/services.  If 
you all do that, and agree on a common port number you will establish a 
common port number that can be used for python code interchange and even 
broadcasting among the group.  Since thats a 4 byte wide hex number, and 
there's 65536 of them, it shouldn't be that hard to find one that will 
serve the purpose.

But I do also,  have a router running dd-wrt facing the net, my guard dog 
with the big teeth.

It (that number) will in time, become that standard.

> You can get a whole virtual server from Amazon for free that is
> perfectly good for hosting git repos and lots of other things too, or
> you can get them for a few dollars a year in other places if you don't
> want to deal with Amazon.  Github adds some conveniences but I've
> never understood the attraction and mystique around it.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

[toc] | [prev] | [next] | [standalone]

#87399 — Re: Code hosting providers

On Fri, 13 Mar 2015 19:38:09 -0400, Gene Heskett <gheskett@wdtv.com>
wrote:

>
>Running your own server is a piece of cake, and if I, at 80 yo, can do 
>it, I don't see a single reason you can't do likewise.  The code I 
>write, for what is called a legacy computer, is just one of the things I 
>share at the link in my sig.  That link is actually _this_ machine.

That's taking things too far. And when people speak of hosting your
own server, they don't necessarily mean hosting in your home computer.
Speaking for myself, I refuse to collaborate on any project that is
hosted on some dude's personal computer. The uptime can never match
that of a data center and and if you decide to stream some movies
while torrenting the latest linux distro, I'm basically screwed.

If you are the sole developer of your project, why not. But if that is
the case, I'd say most project managers are just overkill and the
choice of what to choose shouldn't even be considered. Just grab your
VCS of choice. Issue tracking can be safely managed in your code with
TODO and FIXME. If by any chance you need to work on more than one
computer, don't bloody open your ports at home. Just get an usb pen
and make a bare repo in there from where you can push/pull.

The ability to host your own server means:

    1. Your code hosting provider doesn't lock you in. You can take 
       your *full* repository history and host it yourself somewhere 
       else outside their domain.
    2. You will want to host this ASAP on some web hosting provider.

[toc] | [prev] | [next] | [standalone]

#87402 — Re: Code hosting providers

Mario Figueiredo <marfig@gmail.com> writes:
> That's taking things too far. And when people speak of hosting your
> own server, they don't necessarily mean hosting in your home computer.
> Speaking for myself, I refuse to collaborate on any project that is
> hosted on some dude's personal computer. 

Meh, you don't necessarily know where it is, but you can get a year of
EC2 VPS for free (https://aws.amazon.com/free/), or for ridiculously
cheap at tons of smaller providers (http://list.lowendserv.net/).  Even
the smallest of such VPS is plenty for a moderate sized Git repo.

[toc] | [prev] | [next] | [standalone]

#87405 — Re: Code hosting providers

On Friday 13 March 2015 20:48:36 Mario Figueiredo wrote:
> On Fri, 13 Mar 2015 19:38:09 -0400, Gene Heskett <gheskett@wdtv.com>
>
> wrote:
> >Running your own server is a piece of cake, and if I, at 80 yo, can
> > do it, I don't see a single reason you can't do likewise.  The code
> > I write, for what is called a legacy computer, is just one of the
> > things I share at the link in my sig.  That link is actually _this_
> > machine.
>
> That's taking things too far. And when people speak of hosting your
> own server, they don't necessarily mean hosting in your home computer.
> Speaking for myself, I refuse to collaborate on any project that is
> hosted on some dude's personal computer.

Chuckle.  1. My finished, running code is copied to at least 200 other 
users machines because it is useful code. 

2.  If I fall over, there is close to that many copies extant that 
include the src code since that is virtually the only way I have 
released anything in the last 25 years.

> The uptime can never match 
> that of a data center and and if you decide to stream some movies
> while torrenting the latest linux distro, I'm basically screwed.

That is much more of a function of the bandwidth available to me, 10 
megbit dl, 2.5 megbit ul.  I, on SS, do not have the budget for a 
gigabit pipe, and I have serious doubts that level of bandwidth is even 
available locally.

> If you are the sole developer of your project, why not.

Generally I am, unless a circuit board is under construction in our 
group, in which case I might offer to make a couple of non-plated thru 
copies if I get to keep one for my time. However, on my small machinery, 
mechanical etching, while great on the environment as there's  no 
downright nasty chemistry to dispose of, is also very slow, limited by 
the 2500 rpms max speed of the spindle, so a busy double sided 4"x8" 
board is over 8 hours to machine.  That big, I'd also have to make a 
vacuum pallet, which for a given sized board is about a day.

> But if that is 
> the case, I'd say most project managers are just overkill and the
> choice of what to choose shouldn't even be considered. Just grab your
> VCS of choice. Issue tracking can be safely managed in your code with
> TODO and FIXME. If by any chance you need to work on more than one
> computer, don't bloody open your ports at home. Just get an usb pen
> and make a bare repo in there from where you can push/pull.

And if the next box you plug it into has a different first user number, 
and you don't have a root account handy, how do you access the data, 
written say on a fedora system, but you need to read it on a debian 
system. And you are not the first user, and you are not in /etc/sudoers.

Go ahead, I'll wait for you to suss that out. :)

Besides, thats way more trouble than NFS shares or ssh -Y sessions on my 
local network assuming I am the first user and therefore in the sudoers 
file..  And with dd-wrt playing guard dog for the whole system, I don't 
worry about somebody stomping in and doing an rm -Rf /.  No one has in 
well over a decade.

But, while I have done quite a few scripts on this machine, the one my 
legacy code runs on is not accessible from the outside world.  The 
kernel modules to support that are not in any of the several choices of 
specialized bootfiles for it.

> The ability to host your own server means:
>
>     1. Your code hosting provider doesn't lock you in. You can take
>        your *full* repository history and host it yourself somewhere
>        else outside their domain.
>     2. You will want to host this ASAP on some web hosting provider.

What for?  I keep backups using amanda.  Because of that, and despite 
several hard drive deaths, my email corpus for some mailing lists is now 
13 yo, and several gigabytes.

Question:  How much money is this group, taken as the whole of the python 
world, spending on remote hosting per month?

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>

[toc] | [prev] | [next] | [standalone]

#87406 — Re: Code hosting providers

On Fri, 13 Mar 2015 23:13:17 -0400, Gene Heskett <gheskett@wdtv.com>
wrote:
>
>>
>> That's taking things too far. And when people speak of hosting your
>> own server, they don't necessarily mean hosting in your home computer.
>> Speaking for myself, I refuse to collaborate on any project that is
>> hosted on some dude's personal computer.
>
>Chuckle.  1. My finished, running code is copied to at least 200 other 
>users machines because it is useful code. 

I said collaborating, not just downloading your code. By collaborating
I mean forking and branching your project, making pull requests,
merging, opening and closing tickets, etc.


>
>> If you are the sole developer of your project, why not.
>
>Generally I am, unless a circuit board is under construction in our 
>group, in which case I might offer to make a couple of non-plated thru 
>copies if I get to keep one for my time. However, on my small machinery, 
>mechanical etching, while great on the environment as there's  no 
>downright nasty chemistry to dispose of, is also very slow, limited by 
>the 2500 rpms max speed of the spindle, so a busy double sided 4"x8" 
>board is over 8 hours to machine.  That big, I'd also have to make a 
>vacuum pallet, which for a given sized board is about a day.

You aren't talking about software anymore... I don't know why you are
telling me this.

>
>> But if that is 
>> the case, I'd say most project managers are just overkill and the
>> choice of what to choose shouldn't even be considered. Just grab your
>> VCS of choice. Issue tracking can be safely managed in your code with
>> TODO and FIXME. If by any chance you need to work on more than one
>> computer, don't bloody open your ports at home. Just get an usb pen
>> and make a bare repo in there from where you can push/pull.
>
>And if the next box you plug it into has a different first user number, 
>and you don't have a root account handy, how do you access the data, 
>written say on a fedora system, but you need to read it on a debian 
>system. And you are not the first user, and you are not in /etc/sudoers.
>
>Go ahead, I'll wait for you to suss that out. :)

What do you mean? If you have those problems with a bare repo in a usb
pen, you will have those problems with a bare repo anywhere, including
in your computer back at home.

How do you propose to solve any local access previligies by having
your project management system hosted at your home? I don't
understand.

...I'm getting the feeling we are talking of different things.

>What for?  I keep backups using amanda.  Because of that, and despite 
>several hard drive deaths, my email corpus for some mailing lists is now 
>13 yo, and several gigabytes.

Right. You won't ever lose your data. Heard that before.

I'm a bit old too, you know. And I learned the following, back in the
late 80s when I was initiating my computer carreer doing nightly
backups on a ES9000:

"You will never lose your data until you lost it."

>
>Question:  How much money is this group, taken as the whole of the python 
>world, spending on remote hosting per month?

I'd wager very little, since most options are completely free.

[toc] | [prev] | [next] | [standalone]

#87407 — Re: Code hosting providers

Mario Figueiredo <marfig@gmail.com> writes:
>>Question:  How much money is this group, taken as the whole of the python 
>>world, spending on remote hosting per month?
> I'd wager very little, since most options are completely free.

Oh come on, if you count all forms of hosting, some of us are spending a
lot (megabucks per month on AWS in some cases I'm sure).  If you just
mean VCS hosting, the free Github plans are only for FOSS projects.
Prices of paid plans vary depending on project size etc.  Lots of
companies pay to use Github for closed source stuff.

[toc] | [prev] | [next] | [standalone]

#87408 — Re: Code hosting providers

On Saturday, March 14, 2015 at 9:15:39 AM UTC+5:30, Paul Rubin wrote:
> Mario Figueiredo writes:
> >>Question:  How much money is this group, taken as the whole of the python 
> >>world, spending on remote hosting per month?
> > I'd wager very little, since most options are completely free.
> 
> Oh come on, if you count all forms of hosting, some of us are spending a
> lot (megabucks per month on AWS in some cases I'm sure).  If you just
> mean VCS hosting, the free Github plans are only for FOSS projects.
> Prices of paid plans vary depending on project size etc.  Lots of
> companies pay to use Github for closed source stuff.

Bitbucket will give you closed source repos for upto 5 users collaborating for FREE.
I use it the way others (I guess) you dropbox/googledrive etc

[toc] | [prev] | [next] | [standalone]

#87366 — Re: Code hosting providers (was: Google Code Shutting Down)

On Fri, 13 Mar 2015 12:37:30 +1100, Ben Finney
<ben+python@benfinney.id.au> wrote:

>
>Any service which doesn't run their service on free software is one to
>avoid <URL:http://mako.cc/writing/hill-free_tools.html>; free software
>projects need free tools to remain that way.
>
>
>GitLab <URL:https://about.gitlab.com/> is a good option: they provide
>VCS, file hosting, wiki, issue tracker, code review via merge requests,
>etc. and all of it can be migrated to any other instance of the same
>service.
>

For sure the Communition Edition offers self-hosting. But in face of
Benjamin article above, what is your opinion on its commercial
approach to provide only relevant features through its vendor lock-in
service?

I always wondered what would happen if the developers community
chooses to take its open-source Community edition and fork it into a
EE edition.

Reminds me so much of MySQL and why MariaDB is today a superior
DBMS...

>Also worth watching is Kallithea, a new federated code hosting service
><URL:https://kallithea-scm.org/>. It supports Mercurial and Git for VCS,
>code review, and integrates with existing issue trackers. Because it's
>federated, you won't suffer from vendor lock-in.
>
>Good hunting in finding a free-software code hosting provider for your
>projects!

And Redmine. Kallithea and Redmine seem like two great options. The
only problem with Redmine is that it does not provide any Git workflow
(correc tme if this is not true anymore).

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | comp.lang.python

csiph-web