The goal here is to focus on the original hardware/software through
AIX 5.2 or so. If you know of a site, files, documents or anything
else please reply here or via email and I will add them as time
permits.
Right now the site is somewhat PS/2 centric but RS/6000 is being
assembled.
The URL is:
http://ps-2.kev009.com:8081/
Also, would there be any interest in a System P wiki page that anyone
could contribute information, specs, history/folklore, etc. to?
PS:
Try the full text search for queries when Google is too polluted with
spare parts suppliers.
Regards,
Kevin Bowling
that's not IBM specific. Some others are even worse.
> I am looking to find sites, files, and information for AIX
> and RS/6000s in danger of disappearing. I've started with some older
> manuals and the AIX 4.3.3 and 5.1 fixpacks.
Well, mirroring their ftp sites "only" requires a lot of disk space,
and plain http sites could in principle be retrieved via webarchive.org,
in case the original has disappeared.
What I consider to become an increasing problem are sites which
dynamically generate their content from some database,
after the user has pressed a gazillion of buttons or selection lists.
E.g. IBMs ancient sales manuals like:
> The goal here is to focus on the original hardware/software through
> AIX 5.2 or so. If you know of a site, files, documents or anything
> else please reply here or via email and I will add them as time
> permits.
rootvg's archive (http://archive.rootvg.net/) comes to mind,
much more interesting stuff than their current main site.
> Right now the site is somewhat PS/2 centric but RS/6000 is being
> assembled.
>
> The URL is:
> http://ps-2.kev009.com:8081/
a well-known site, I might appreciate it even more,
whenever I'm trying to revive my RT-PC in the basement :-)
Have you looked at www.rootvg.net?
Just in case you really responded to me and not the OP:
yes I did, but rootvg addresses current topics
(which is appreciated of course) and
many of the links just point to IBM sites.
And it's full of annoying cookies :-(
But the OP's intention was to preserve older docs
before they go away for good.
> Well, mirroring their ftp sites "only" requires a lot of disk space,
> and plain http sites could in principle be retrieved via webarchive.org,
> in case the original has disappeared.
The idea is to somewhat intelligently select the content.. I don't
think sucking down all of IBM's FTP would be very helpful since there
is a lot of redundancy and cruft.
High on my priorities is identifying fixpacks/rollups for AIX 3.x.x
and 4.1-4.2 since some machines like the PPC laptops cannot use newer
ones. I don't have any experience with these versions though, so
could use some help finding what to slurp from IBM FTP. Also doing
the same for some of the common VisualAge C++ versions.
The Internet Archive Project is a last resort in my opinion. It is
slow and not very thorough. I've also already found my full text
search immensely useful.
> What I consider to become an increasing problem are sites which
> dynamically generate their content from some database,
> after the user has pressed a gazillion of buttons or selection lists.
I've noticed that a lot on IBM requires you to have a verified
business account as well.
>
> rootvg's archive (http://archive.rootvg.net/) comes to mind,
> much more interesting stuff than their current main site.
Yes, I have this. There isn't really a lot of information on that
page anyways. Most of it is outdated linkage to IBM. The now defunct
s6000.com page (now just a shell for a parts supplier.. not that the
original page was much else) isn't even worth digging out of archives.
such as my ThinkPad 860, up and running networked with 4.1 :-)
Only the accu seems to be dead :-(
> cannot use newer
> ones. I don't have any experience with these versions though, so
> could use some help finding what to slurp from IBM FTP. Also doing
> the same for some of the common VisualAge C++ versions.
has a lot of such old stuff.
74+ GB.
> The Internet Archive Project is a last resort in my opinion. It is
> slow and not very thorough.
That's true, unfortunately.
What I miss so far is a functionality such as
"give me a complete off-line snapshot of the searched site",
i.e. sth like "wget" with a time specifier.
http://ps-2.kev009.com:8081/rs6000/
Lots of new stuff: RSINFO/6000 (html version), RS/6000 Technical
Library, new PDFs.
Also, I can't seem to download the newer RSInfo from IBM's wiki:
www.ibm.com/developerworks/wikis/download/attachments/52133983/RSinfo.pdf?version=1
It gets stuck on 91% and corrupt no matter the computer or network I
use.
> ftp.software.ibm.com/aix/
>
> has a lot of such old stuff.
> 74+ GB.
Again, I don't think it would be terribly useful to mirror the whole
thing. There is a lot of junk, duplicates, and stale data I.E that
has been merged into MLs or rollups. I will go through this manually
next week and see if I can't pick out things I think are useful. If
you have any suggestions, let me know!
I've launched a wiki and forum as well. No annoying ads!
http://www.ibmhardware.net/wiki/Main_Page
http://www.ibmhardware.net/forum
The goal of these pages is to cover new hardware as well. I need help
finding some good moderators and editors.
Regards,
Kevin