On 07/18/2010 05:50 PM, Paul Menzel wrote:
[ Please reply in context. ]
Am Sonntag, den 18.07.2010, 17:36 -0400 schrieb Gerry Reno:
[…]
The whole point of tool discussion in an open source project should be
what tools make it easiest for the community as a whole to contribute,
collaborate and participate in the project.
If you have two main developers, I would take that into consideration.
And since there is no big community yet, nobody knows what the community
is going to prefer.
As the community grows it will be composed of "open source" participants
and contributors as it always is. And these people are already familiar
with Launchpad and SourceForge for the most part because they have been
almost always participating and contributing in other open source
projects which on average are hosted on one of those two forges.
When you look at the massive success of forges such as Launchpad and
SourceForge that didn't happen accidentally. It happened because these
forges are extremely good at providing all of the infrastructure,
collaboration tools, tracking tools, and source code management tools
for an open source project.
I also heard of some projects moving away from those services.
And that is not recently. Yes, there were some capacity issues on the
major forges for a while when they got entirely overwhelmed by their
success. Those issues have been resolved for quite a while now. And
recently you are seeing really big-name projects such as MySQL, MariaDB,
and GNU projects moving onto Launchpad.
To try and duplicate this type of free services using your own hardware,
software, and infrastructure is just not possible. Why go back and use
'bone knives and bearskins' with your own infrastructure and waste
countless valuable hours reinventing the wheel when you can have all
this handed to you on a silver platter for free? I think this is why
hundreds of thousands of projects are using the forges and spending
their time thinking about their projects instead of all the necessities
of supporting an infrastructure.
As I have already written.
1. Independence. There is the danger that you are locked into the
project and depend on their reaction time. See the BerliOS and renaming
the mailing list issues. Using a server for yourself you are in control
of it.
In control of a single point of failure. Great. And you could buy an
electrical generator and make your own electricity. And maybe a wafer
production facility and make your own hardware chips.
2. As I have written. Maybe it is no overhead for Alex or Heinz since
they need to administer a server and those services anyway. We both do
not know that.
That's their choice if they want to fiddle around with hardware. But
the project should not be based on developers 'single point of failure'
server. It needs to be placed at a forge service provider that has all
the inhouse expertise to deal with hosting open source projects and the
required infrastructure and services.
Regards,
Gerry
_______________________________________________
X2go-dev mailing list
X2go-dev@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/x2go-dev
