[LMH]Exploiter-9d0820

[James A. Crippen]

Nyef <nyef@softhome.net> writes:

> On Thu, 21 Aug 2003, Robert Swindells wrote:
>
>> Alastair Bridgewater wrote:
>> 
>> >This version requires a disk image file as well as a load band.
>> 
>> Is there a suitable disk image available somewhere ?
>> 
>> Robert Swindells
>
> Umm... No? And definately not the explScsi0Dr0.dsk file in the archive at 
> http://sc.am/~hamethief/lispm/explorer.tar (which is also not a copy of 
> the same archive that Al Kossow supplied), which wasn't mentioned on the 
> list recently.

This brings me to mentioning what I've been thinking about recently: a
good location or even some mirrors for these rather large files that
we're dealing with nowadays. First it was just a load band, but now
various bands and disk images are necessary. I don't really want to
host them all since I have a monthly bandwidth cap of 10 GB that could
be easily chewed up by lots of people sucking down copies, and
secondly because I'm in Alaska, which isn't particularly close to
*anywhere*, network-topologically. Thirdly, I only have a tiny straw
of 64 kbps outbound through which to suck.

So maybe we should come up with some structure for the set of
'original' data files that are useful for testing, and then get
together some sort of mirror group so that people in Europe can get
them in one location, people in the US at another, and freaks close to
Alaska (Washington, Oregon, and BC residents?) can get them from my
site.

Any offers of space and bandwidth? Particularly we'll need at least
one Europe site as well as one in the contiguous 48 states (or in
neighboring Canada, I suppose).

Let me look at all the files I have gathered and I'll see about
organizing them into a coherent package. As for mirroring, I suppose
the easiest thing is just to have the mirror owners get new additions
from the source site (me?) as they are announced on the list.

A side benefit is that we can start to collect donated bands and disk
images from other Explorer owners and archive them somewhat sanely.
Then our statistical sample of real test data will grow and maybe make
the code more realistic.

'james



BTW, I'm sure my network connectivity situation will improve
dramatically when I finally move to Honolulu in a couple years... Not!

-- 
James A. Crippen <james at unlambda.com> Lambda Unlimited
61.2204N, -149.8964W                     Recursion 'R' Us
Anchorage, Alaska, USA, Earth            Y = \f.(\x.f(xx))(\x.f(xx))