[Bitcoin-development] Block Size Increase

Fri May 8 17:17:49 UTC 2015

On Fri, May 8, 2015 at 2:59 PM, Alan Reiner <etotheipi at gmail.com> wrote:

>
> This isn't about "everyone's coffee".  This is about an absolute minimum
> amount of participation by people who wish to use the network.   If our
> goal is really for bitcoin to really be a global, open transaction network
> that makes money fluid, then 7tps is already a failure.  If even 5% of the
> world (350M people) was using the network for 1 tx per month (perhaps to
> open payment channels, or shift money between side chains), we'll be above
> 100 tps.  And that doesn't include all the non-individuals (organizations)
> that want to use it.
>

> The goals of "a global transaction network" and "everyone must be able to
> run a full node with their $200 dell laptop" are not compatible.  We need
> to accept that a global transaction system cannot be fully/constantly
> audited by everyone and their mother.  The important feature of the network
> is that it is open and anyone *can* get the history and verify it.  But not
> everyone is required to.   Trying to promote a system wher000e the history
> can be forever handled by a low-end PC is already falling out of reach,
> even with our miniscule 7 tps.  Clinging to that goal needlessly limits the
> capability for the network to scale to be a useful global payments system
>

These are good points and got me thinking (but I think you're wrong). If we
really want each of the 10 billion people soon using bitcoin once per
month, that will require 500MB blocks. That's about 2 TB per month. And if
you relay it to 4 peers, it's 10 TB per month. Which I suppose is doable
for a home desktop, so you can just run a pruned full node with all
transactions from the past month. But how do you sync all those
transactions if you've never done this before or it's been a while since
you did? I think it currently takes at least 3 hours to fully sync 30 GB of
transactions. So 2 TB will take 8 days, then you take a bit more time to
sync the days that passed while you were syncing. So that's doable, but at
a certain point, like 10 TB per month (still only 5 transactions per month
per person), you will need 41 days to sync that month, so you will never
catch up. So I think in order to keep the very important property of anyone
being able to start clean and verify the thing, then we need to think of
bitcoin as a system that does transactions for a large number of users at
once in one transaction, and not a system where each person will make a
~monthly transaction on. We need to therefore rely on sidechains,
treechains, lightning channels, etc...

I'm not a bitcoin wizard and this is just my second post on this mailing
list, so I may be missing something. So please someone, correct me if I'm
wrong.

>
>
>
> On 05/07/2015 03:54 PM, Jeff Garzik wrote:
>
>  On Thu, May 7, 2015 at 3:31 PM, Alan Reiner <etotheipi at gmail.com> wrote:
>
>
>>  (2) Leveraging fee pressure at 1MB to solve the problem is actually
>> really a bad idea.  It's really bad while Bitcoin is still growing, and
>> relying on fee pressure at 1 MB severely impacts attractiveness and
>> adoption potential of Bitcoin (due to high fees and unreliability).  But
>> more importantly, it ignores the fact that for a 7 tps is pathetic for a
>> global transaction system.  It is a couple orders of magnitude too low for
>> any meaningful commercial activity to occur.  If we continue with a cap of
>> 7 tps forever, Bitcoin *will* fail.  Or at best, it will fail to be
>> useful for the vast majority of the world (which probably leads to
>> failure).  We shouldn't be talking about fee pressure until we hit 700 tps,
>> which is probably still too low.
>>
>  [...]
>
>  1) Agree that 7 tps is too low
>
>  2) Where do you want to go?  Should bitcoin scale up to handle all the
> world's coffees?
>
>  This is hugely unrealistic.  700 tps is 100MB blocks, 14.4 GB/day --
> just for a single feed.  If you include relaying to multiple nodes, plus
> serving 500 million SPV clients en grosse, who has the capacity to run such
> a node?  By the time we get to fee pressure, in your scenario, our network
> node count is tiny and highly centralized.
>
>  3) In RE "fee pressure" -- Do you see the moral hazard to a software-run
> system?  It is an intentional, human decision to flood the market with
> supply, thereby altering the economics, forcing fees to remain low in the
> hopes of achieving adoption.  I'm pro-bitcoin and obviously want to see
> bitcoin adoption - but I don't want to sacrifice every decentralized
> principle and become a central banker in order to get there.
>
>
>
>
> ------------------------------------------------------------------------------
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
> _______________________________________________
> Bitcoin-development mailing list
> Bitcoin-development at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>
>

-- 
PGP: B6AC 822C 451D 6304 6A28  49E9 7DB7 011C D53B 5647
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20150508/b5114625/attachment.html>