[ih] reinventing the wheel, was Internet History Lives on the Internet?

[Jack Haverty]

OK, I'm admittedly illiterate at this BitTorrent Wheel technology.   I'd
be happy to hear that it is the solution to historical archiving.   I'm
willing to set up a BT client and I can probably figure out how to do it
and learn.  Wikipedia says that BT had about 150 million active users,
with 20 million concurrent, and that statistic was at least 5 years
ago.  So I guess I'm coming pretty late to that party.

But I've never used these BT Wheels, so perhaps someone who has been an
active BT user can help me understand how to do it right.  

I'll make this concrete.

It is 2019, the 50th anniversary of the ARPANET (which IIRC is what
started this discussion).   I was part of the crew who put together the
"coming out party" for the ARPANET, held in 1972 at a conference with a
live demo in a ballroom at the Washington Hilton. 

In my basement I still have a folder containing materials I brought home
from that event.  For example, I have one of the brochures which was
handed out to all attendees - the ARPANET Demo brochure, which contains
things like the floor plan of the demo layouts in the ballroom.   It
seems like that might be of interest to some people during the 50th
anniversary.   So I'd like to make it available.  Scanning it produces a
few MB of image file.

About 4 years ago, I scanned that brochure and emailed it to a
professional historian who was looking for ARPANET historical
materials.  He was somewhat ecstatic and said that he had never seen
that document before and hadn't found it online despite extensive
searching.   Googling it now doesn't seem to find it.   But perhaps he
put it up on BitTorrent (assuming that's the proper terminology), and I
just don't know how to spin that wheel.

So, once I get a BT client running, I'll try to make sure that artifact
is available.  I'm assuming a BT client has some kind of "search"
capability, so I can check to see if it's already there.  If not, I'll
upload it (or whatever) to BT.  I'm pretty sure I can figure it out.

So, here's my questions:

1- How long after uploading do you typically have to wait before a
seasoned BT user somewhere else in the world can search and find that
brochure file and download it?   When can I say "It's on BitTorrent, go
get it there"?

2- For all the BT-illiterate people like me, how long will it typically
be before the common search engine spiders crawl through the new BT
content and find that brochure file, so that a Google/Bing/etc search
will produce it in their search results for downloading?  When can I say
"Just Google it."

3- How long do I have to keep my computer with the BT client running
before I can turn it off, and be confident that the brochure has gotten
into enough other machines that it will remain accessible?  Is there a
way for me to tell it's safe to do so?  Is it Hours?  Days?  Months?  Years?

4- How long does material survive within BT if nobody downloads it? 
ARPANET artifacts may be popular during 2019, but much less so in the
51st anniversary year.  Very few people may look for them for a long
time.  Most computers running today (and their owners) will be long gone
when the 100th anniversary of ARPANET revives interest in such things. 
It seems likely that the Internet will still be around, even still
hosting a parade of BT-enabled computers that has been coming and going
continuously for 50 years.  When the AI that's planning the 100th
anniversary festivities notices a mention of the 1972 ARPANET brochure
in the writeups of the 50th anniversary, will it still be able to find
that brochure file on BT in 2069 and download it?  

That's what I mean by a Persistent Store... an Internet-based system
that stores data continuously, regardless of how frequently or widely it
gets used, and how many times the physical computers and storage devices
and people come and go, as long as there are always enough computers and
storage in the system at any time. (Like a BotNet, or the PARC worm). 
People who are interested in History can donate their excess computer
time and memory to "Help Save History" by downloading some software
(like SETI).

If that's what BT already does, all we all have to do is put all our old
stuff into it.   I'm still skeptical but I'll try BT and learn about
it.  It is pretty clever technology, but my gut feeling is that there's
more needed.  I suspect, hope, that perhaps some clever mashup of the
technology already in BT, Botnet, Blockchain, and the like would do the
trick.

/Jack



On 2/25/19 6:46 AM, John R. Levine wrote:
>>     - Content is stored outside of BitTorrent per se, e.g., on web
>> servers.  Torrent servers essentially act as caches for material in
>> transit. ...
>
> Dunno where you'd have heard that, because that's exactly what
> bittorrent and other p2p software don't do.  Every node serves what it
> has, including material it's downloaded from other nodes.  Each file
> is divided into smallish pieces which are downloaded and served
> separately.  If you're downloading, say, a linux DVD ISO, which is a
> thousand pieces, you can watch as it sucks down and reassembles the
> pieces in whatever order it can find them, and at the same time starts
> serving the pieces it's already got.  If you leave the client running,
> it'll continue to serve them.
>
> One of the reasons that people use p2p for pirated stuff is that
> there's no central copy, so there's no single place one can make a
> file go away. It really is your file botnet.
>
>>     - There isn't any obvious (to me) way to restrict what your personal
>> machine might store or serve.
>
> It only serves what it's got, and it's only got what you've told it to
> download or added locally.
>
> Since you are apparently interested in p2p file distribution, why not
> install a bittorrent client (the ones at bittorrent.com are OK) and
> try it out for a few minutes.
>
> R's,
> John
>
> PS: I realize this is the Internet History list, but it's like the
> history stopped 20 years ago.