[kwlug-disc] Callout to CS majors/Other technically savvy klwug listers.

Stuart Seeley stuart at lowlevel.ca
Fri Feb 8 15:17:58 EST 2019


Blockchain seems to be applied to almost everything these days and that
doesn't usually make a lot of sense, but I think it does in LBRY's case.
The payment aspect of it to me is very much secondary to the
anti-censorship...  the end users host the index on blockchain and the
content on torrents; can be available indefinitely with no server overhead.

With typical torrent tracker/indexing site.. or content hosting sites...
would be consumers of data can easily be cut off from content since there
is someone in control of the central point deciding what can be available
and what can't be.  (Or authorities deciding which trackers or hosts can
exist and can't.)

I have some concerns that LBRY may have trouble with 51% attacks, or out of
control blockchain data growth/etc, but perhaps they have that figured out.
The bitcoin block chain is pushing 200 gigabytes.

Stuart


On Thu, Feb 7, 2019 at 8:38 PM Doug Moen <doug at moens.org> wrote:

> Setting up a decentralized digital content marketplace is on my TODO list
> for this year. I want a way for users to create and share content, and find
> and access other user-created content, in conjunction with my open source
> project Curv, which allows you to design 3D models for 3D printing. This
> hypothetical network of shared content could be called "CurvNet".
>
> Creating a payment system isn't part of my plan, though, since it could
> incentivize users to put their content behind a paywall, and prevent
> share/reuse/remix open source/open culture behaviour. It's the difference
> between thingiverse.com (a free/open site for uploading/downloading 3D
> digital models, with no payment model) and shapeways.com (you upload your
> models then charge people to get 3D printed copies of your models. It's
> theoretically possible to offer your models for free download, but in
> practice nobody does). Ted Nelson's project Xanadu promised a way to pay
> authors while incentivizing sharing/reuse/remix, because in his system, the
> more people reuse/remix your content, the more you get paid. That would be
> ideal, but I'm not aware that anybody knows how to actually implement this
> on today's internet.
>
> The centralized model (eg, how github works) has the advantage of being
> well known, it's obvious how to set it up. The problem is that I don't want
> to pay for the servers needed to keep the CurvNet running, especially if it
> the project takes off and becomes really popular. Creating a business model
> that pays for the servers to keep running is also problematic. It's either
> advertising (collecting people's personal information and reselling it), or
> you pay for an account, which creates a large barrier to entry. Plus it's
> centralized, so if the CurvNet server shuts down, then everybody loses.
>
> I'd prefer a decentralized model. But where are the servers where the
> content is stored; who pays for those servers? How do you find content on
> the network related to your interests? How much friction is involved in
> publishing content to the network?
>
> Do you need to install an application on your laptop/desktop/tablet in
> order to start using the application, or is it as easy as visiting a URL
> and have the content browser/content editor application immediately start
> running in your browser? Installing an application creates friction, but
> then it provides an answer to where you store your personal content: on
> your laptop/desktop/tablet's local storage. Having the application run in a
> web browser, and storing your personal content on the cloud, reduces
> friction, and lets you access your content from any device. But as I asked
> early, where are the cloud servers and how are they paid for?
>
> LBRY uses blockchain. That's a red flag:
>
> https://www.wired.com/story/theres-no-good-reason-to-trust-blockchain-technology/
> The author, Bruce Schneier, is an expert in the field of security and
> trust who is worth listening to, and his arguments are good. Any blockchain
> based solution should address the points that Schneier makes.
>
> Anyway, that is my starting point. I will look at LBRY and see if there
> are any new answers to my questions.
>
> Also, I would definitely attend a KWLUG talk that provided new answers to
> these questions.
>
> Doug Moen.
>
> On Thu, Feb 7, 2019, at 10:31 AM, Chamunks wrote:
>
> Hey KWLUG,
>
> My company, LBRY, just finished a beta draft introducing its technology.
> We're looking for feedback from the best and brightest.
>
> [You can check it out at https://lbry.tech/spec](https://lbry.tech/spec).
> The intended audience is computer scientists and/or professional software
> engineers.
>
> Any thoughts you have at all would mean a lot to me and our entire team.
> Is it interesting? Does it make sense to you?
>
> P.S. You're welcome to share the link with others, but we're looking to
> avoid posts to any public forums (for now). I may contact you again in a
> few weeks when we're ready for a more public campaign.
> *_______________________________________________*
> kwlug-disc mailing list
> kwlug-disc at kwlug.org
> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org
>
>
> _______________________________________________
> kwlug-disc mailing list
> kwlug-disc at kwlug.org
> http://kwlug.org/mailman/listinfo/kwlug-disc_kwlug.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://kwlug.org/pipermail/kwlug-disc_kwlug.org/attachments/20190208/3f592468/attachment-0001.html>


More information about the kwlug-disc mailing list