Re: [Mingw-users] Is MinGW GCC 4.5.0 stable or not?

[Keith M. <kei...@us...>]

On Tuesday 04 May 2010 10:13:58 Tao Wang wrote:
> However, regarding the "4.5.0 is a stable release", um... , MinGW
> offcial website cannot be a 'reliable source', because I cannot
> find anything directly state that 4.5.0 is a current stable
> release in the website. So, unless you update your website or
> public document to make it clear, it's hard likely to be a
> reference.

I really didn't want to get drawn into this futile discussion, so 
I'll just say this, and let it be an end to it:

http://mingw.org/PackageIdentificationHOWTO
|Release Status
|
|Generally, this field is used only for packages that undergo 
|extensive, and iterative, development by the MinGW team, outside the 
|confines of the upstream project. The prime example here is the 
|elements of the MinGW GCC packages themselves. There are several 
|possible values:    
|
|   * alpha: Unstable, work in progress. Probably has many known and 
|     unknown sharp edges; you will get cut. 
|   * beta: Unstable, but suitable for most users to begin 
|     experimenting. Don't use it for production code. 
|   * rc: Release candidate. Ready for general use, and needs 
|     community involvement to flush out and fix any last minute 
|     bugs.  These will probably be minor.  
|   * stable: Final production release. Most stable releases will  
|     omit the Release Status field entirely; thus, it follows that 
|     any current package which omits this qualifier is considered to 
|     be a stable release.   

Can't really make it much clearer than this; the gcc-4.5.0 release 
tarball omits the "release status" qualifier, ergo it is a stable 
release.  (Yes, I did add an explicit clarification to this effect, 
prompted by this thread; however, this has been implicit in the 
foregoing statement, since Chuck and I -- predominantly Chuck -- 
wrote the original page content).

Also note that this page, unlike the others cited, is *not* amongst 
the wiki pages on mingw.org.  Thus, its content should be better 
controlled, and therefore of more reliable provenance than the wiki 
pages.  (A wiki is only as reliable as the comprehension of those 
who contribute -- and this is equally true of our wiki and of 
Wikipedia.  While we are grateful to those who do contribute, we 
simply don't have the resources to audit it all, so errors and 
potentially misleading content will inevitably creep in).

-- 
Regards,
Keith.