[p2p-research] what about archive.org as an alternative to youtube?

[Samuel Rose]

On Wed, Dec 22, 2010 at 4:33 AM, Dan Brickley <danbri at danbri.org> wrote:
> On Wed, Dec 22, 2010 at 3:54 AM, Samuel Rose <samuel.rose at gmail.com> wrote:
>> We used to have some friends that recorded our live music shows and
>> uploaded to archive.org
>>
>> Archive.org could use some help in the coming years creating a way to
>> massively distribute it's archive, in my opinion. They will eventually
>> run out of space and resources on their existing servers. Other than
>> that  theoretical issue, I think archive.org personally is an awesome
>> place to post videos (would still allow you to embed as easily as
>> youtube does)
>
> With storage costs dropping still, even though it's counter-intuitive,
> centralisation can be quite attractive.
>

> There's a Brewster Kahle talk on video up there somewhere, where he
> says something like that you could have textual scans of every book
> ever written in a server farm that costs less than the price of a
> house.


This is true. I say him give this Open Library talk at University of
Michigan.  I think they have around 5-7 industrial grade servers
around the world right now. You are right that it is practical to do
what archive.org is doing now.

The idea for creating a distributed Archive.org actually came from a
discussion we had with Brewster after his talk. Turned out was already
interested in creating some form of distributed storage for
archive.org data before we brought our ideas up to him. I think we're
still going to try and experiment with this in the coming years (a
distributed storage that is limited to a community curated archive,
where participants share part of their hardware to store).

>And talking with TV colleagues yesterday, that a very major
> national TV archive (albeit in low-ish resolution) would fit in a
> not-so-intimidating number of terrabytes.
>
> So in general I'm all for distributed, decentralised, replicated and
> all that. But it is very easy to forget the benefits that come with a
> centralised approach, and also the amazing things that are possible as
> technologies get cheaper. And of course it's not an either-or
> situation. You could have a few archive.org-like supernodes with
> different focus, collections etc, and still build a p2p distribution
> system.
>

You have me thinking: I wonder what it would be like to distribute the
data, then have many less-than super nodes (like simple serving
applications that people can host on lower cost webhosting and/or VPS)
that act as the first serving endpoints?

> Anyhow if you're interested in doing things with TV content from
> archive.org, I recently blogged a howto. Their metadata can be crawled
> but it's not so obvious without help. Some of the archive.org staff
> helped me understand what to do, so I stitched their email
> explanations together to make this post:
> http://danbri.org/words/2010/10/27/565
>

Nice! Thanks a lot.


> cheers,
>
> Dan
>
> _______________________________________________
> p2presearch mailing list
> p2presearch at listcultures.org
> http://listcultures.org/mailman/listinfo/p2presearch_listcultures.org
>



-- 
--
Sam Rose
Future Forward Institute and Forward Foundation
Tel:+1(517) 639-1552
Cel: +1-(517)-974-6451
skype: samuelrose
email: samuel.rose at gmail.com
http://forwardfound.org
http://futureforwardinstitute.org
http://hollymeadcapital.com
http://p2pfoundation.net
http://socialmediaclassroom.com

"The universe is not required to be in perfect harmony with human
ambition." - Carl Sagan