On 09/27/2010 06:37 PM, Finn Thain wrote:
On Mon, 27 Sep 2010, Justin P. Mattock wrote:
On 09/27/2010 09:03 AM, Joe Perches wrote:
On Mon, 2010-09-27 at 11:10 -0400, John W. Linville wrote:
On Sun, Sep 26, 2010 at 11:31:15AM -0700, Justin P. Mattock wrote:
Below is an updated patch from the original fixing broken web addresses in the kernel. Thanks for all the help and info on this to everybody.. Hopefully I didnt miss any of them(if so let me know, and I'll resend).
Changing a URL for a relocated page is one thing, but removing links isn't necessarily a great idea. Even if the site is technically gone, it may be possible to find information e.g through the Internet Archive Wayback Machine.
Perhaps it'd be better to scrape the contents of the various web pages, collect them somewhere like wiki.kernel.org and encourage others to put new contributions in that site.
The copyright problem aside, this might be a good idea for material not already archived but I don't think it makes sense to start a new archive when archive.org (or other) has the information.
And which version(s) do you scrape? I discussed some problems with changing URLs in another thread: http://lkml.org/lkml/2010/9/22/22
Anyway, without knowing what future archive(s) would be available or relevant to any given URL in the future, I think the best we might do is a "Retrieved on YYYY-MM-DD" qualification for new URLs.
yeah I think somebody was saying something about having a separate file, with all the web addresses in them or something...In any case, up to you guys..
I don't see how moving the addresses would help. And would it not make the information harder to find?
Finn
depends on how the setup is.. I was thinking of having some kind of letter/number scheme i.e. (A-13) in the comment, then in the file with all the addresses under article "A" 13th address down you would have the web address.(but keep in mind that's just speculation..)
Justin P. Mattock