[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ih] Internet History Lives on the Internet?

The whole question of persistent storage remains an unsolved problem.

There have been models of distributed publication - like oceanstore and 
publius (huge, distributed hash tables) - but they tend to fall down if 
lots of people don't keep maintaining disk space.

I keep thinking of the notion of a federation of storage providers where 
one pays once for either a block of replicated storage, or for 
publication of a file/document.? These days, a 1TB disk costs $100 
(retail) - so 10cents/GB.? Multiply by 5 for replicated copies, and 
assume a 2-year disk life, and we're talking 25cents/year for a Gig of 
reliable storage (leaving out networking costs).? $25, at 1% interest, 
would "endow" a Gig of reliable storage, "forever" (think about how we 
pay for perpetual care of a gravesite.

What's missing is a legal & accounting mechanism for handling the 
money.? Folks pay to self-publish an e-book - it sure would be nice to 
be able pay, say $50, once, to make a document available for the life of 
the Internet.


On 2/24/19 1:39 PM, Jack Haverty wrote:
> [Changed the subject line]
> I read the recent messages on the forum just before going to sleep, and
> then I had a dream....literally.
> There's a whole different perspective on Internet History that might be
> very revealing.? Instead of questions like "Who built the Internet?",
> perhaps also ask "Who paid for the Internet?"? If historians "followed
> the money" like many other investigators, they might find some
> interesting insights.? I didn't realize until today that the IETF is
> funded by ... Me!?? Through my payments for my .org domain, maybe by now
> I've paid for an urn or two of coffee at an IETF meeting.
> But my dream was of how to fund some kind of Internet repository of
> historical materials, not subject to the management whims or financial
> success of an "institution".?? My dream reminded me that such mechanisms
> already exist, have been running at scale for years, are self-funded,
> and seem essentially impossible to excise even when governments or
> industry giants try to do so.
> My dream is of a Benevolent BotNet (apologies to my alma mater, BBN).
> Instead of hosting and propagating malware and viruses, or stealing
> computer cycle to mine cryptocurrency, the BBN would simply store,
> replicate, and distribute historical materials on demand.? No doubt
> Richard's comment on Pirate Bay triggered this part of the dream.
> Such technology obviously exists, and survives despite serious efforts
> to eradicate it.? Where the Internet was coopted for evil, perhaps the
> evil could be coopted for good?
> Maybe even better would be a mechanism that didn't rely on theft and
> subterfuge at all.? Perhaps something akin to the SETI mechanisms, where
> people voluntarily donate their computer resources to analyze radio
> signals, by simply downloading a piece of code and allowing it to run on
> their computers.
> So, my dream was that some new software appears, which is freely
> downloaded by thousands or millions of people around the world, which
> uses a few GB of the disk on their machines, and stores historical
> material in a redundant, highly survivable, persistent, distrubuted
> historical warehouse.?? One, or many, search engines (go Google!, Bing!,
> DuckDuckGo!) would allow people to find material in the warehouse.
> Anyone could contribute material to the historical archive by simply
> placing a copy into the disk area of their machine that they've shared,
> from where it would be automatically distributed and replicated.
> Perhaps this is one or more apps that can be downloaded.? Or perhaps a
> plug in or extension to popular browsers.? Or maybe an addition to
> existing mechanisms like BitTorrent.? Much of the code already exists,
> as evidenced by the millions of computers unwittingly participating in a
> Botnet, or willingly running code like SETI.
> Dave's offer of disk space is just the start.? I suspect many people
> would contribute some unused chunk of their computers and network
> capacity.? I have a few Terabytes on my NAS that are empty...you
> probably do too.?? With enough participants, the BBN becomes
> self-suficient even as people come and go.
> All it would seem to take is for someone to sit down and write the
> code....in the classic Internet spirit of Rough Consensus and Running Code.
> Dave....?
> /Jack Haverty
> On 2/24/19 7:42 AM, Dave Taht wrote:
>> Joe Touch <touch at strayalpha.com> writes:
>>> On Feb 23, 2019, at 12:42 PM, Jack Haverty <jack at 3kitty.org> wrote:
>>>      But "internet-history at postel.org", and others like it, even RFC
>>>      repositories, likely exist at the whim of their sponsor.
>>> Indeed - even assuming volunteers run them - they?re?s still the issue
>>> of hosting and net access.
>>> I have old repositories (end2end-interest, for one) that even the ISOC
>>> has declined to host (even though the E2E-RG originated there).
>>> Then again, if you want to see the worst of ?free riders?, go attend
>>> an IETF. Companies send armies there for free training and free
>>> consulting.
>>> PS - speaking as list admin, if anyone wants to offer a place to host
>>> this list more reliably and archivally, please do let me know (contact
>>> me directly off-list).
>> My email list server currently lives on linode in the cloud. The cost is
>> $5/month for 25GB of SSD storage. ( https://www.linode.com/pricing
>> ). Has IPv6 and IPv4. It's paid for via a patreon donation.
>> It's not like I'm using much of that box - or the bandwidth available -
>> how big are these archives?
>> I wouldn't mind sharing that existing list server, but I long ago
>> switched to violating whatever RFC it was that said starttls was a
>> "should" - to *mandate* starttls only - which cuts down on spam (and
>> sigh, about 13% of my measured potential correspondents, still). The
>> biggest administrative cost I'd had was dealing with spam.
>> If that's not an acceptable policy for these lists/archives, well, go
>> burn the 5 bucks/mo on yer own.
>>> Joe
>>> _______
>>> internet-history mailing list
>>> internet-history at postel.org
>>> http://mailman.postel.org/mailman/listinfo/internet-history
>>> Contact list-owner at postel.org for assistance.
> _______
> internet-history mailing list
> internet-history at postel.org
> http://mailman.postel.org/mailman/listinfo/internet-history
> Contact list-owner at postel.org for assistance.

In theory, there is no difference between theory and practice.
In practice, there is.  .... Yogi Berra