Subject:
|
Re: NQC beta test release policy
|
Newsgroups:
|
lugnet.robotics.rcx.nqc
|
Date:
|
Sat, 8 Jan 2000 04:52:26 GMT
|
Viewed:
|
1523 times
|
| |
|
|
I'm combining two replies in the thread into a single message...
In article <FnytME.9En@lugnet.com>, Leonard Stiles <lstiles@hotpop.com> wrote:
> My point was about the fact that the
> beta release is not open source, which,
> IMHO, somewhat defeats the idea of
> using a Free Software licence, one of
> the main intentions of which is to
> provide more stable final releases.
I agree completely that philisophically a beta release of free software
should also be free software. I have been cutting corners on this with
NQC out of praticality. There is a non-zero cost to making the source
release in its current form (perhaps 30 minutes of time and an average 2
day delay for the release).
The Free Software model can be used to acheive different goals. In the
case of NQC, the goal is not to make NQC more stable - it is to allow it
to spread to more people and platforms than I could personally support.
NQC's stability comes from other practices.
In article <Fnyx5B.ys@lugnet.com>, Leonard Stiles <lstiles@hotpop.com> wrote:
> About the source release being "tied to the linux release":
> Considering the
> fact that the linux binaries are built from the source, isn't it more the
> case that the linux release is tied to the source release, rather than vice
> versa?
The only reason I consider the source release tied to the linux release is
that there are two source files that only get exercised in the Linux
build. Without building and running the linux version I cannot even
promise that the source will build for other users. In terms of process,
this means that the Linux build is done before I'm willing to "sign off"
on the source (actually all three builds have to take place). Packiging
up the linux binary could be done at a later point, but actually once the
source is on a Linux box and I'm sitting in front of the terminal, a
single script builds both the source and binary Linux packages. However,
just because that's the way I have been doing the release, it doesn't mean
the process can't be changed...
From my end, if I ditched the Linux/x86 executable, then I could do the
entire release on my Mac (including a verification of the Unix source
using LinuxPPC). Would this be a good trade to make...adding source
releases for beta, but dropping linux binary releases entirely?
If there was huge demand for a Linux binary, I suspect there would be
other users willing to build and post it. One user has already sent me a
spec for an NQC RPM, but I haven't had time to work with it. This is the
sort of thing I'd love to have someone else take over since I'm not much
of a Linux user myself.
Dave Baum
--
reply to: dbaum at enteract dot com
|
|
Message has 1 Reply:
 | | Re: NQC beta test release policy
|
| (...) goals. In the (...) is to allow it (...) personally support. (...) Ok, what I meant is that open source beta (and other) releases improve the quality of the overall product. I don't think nqc has a stability problem. (...) entire release on my (...) (25 years ago, 8-Jan-00, to lugnet.robotics.rcx.nqc)
|
Message is in Reply To:
| | Re: NQC beta test release policy
|
| First of all, I don't think there's anything wrong with not releasing binaries of a beta release, be it for linux or any other platform. Of course, if a large sector of the user base is on mac and windows and therefore cannot build their own (...) (25 years ago, 7-Jan-00, to lugnet.robotics.rcx.nqc)
|
9 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
