Very useful.

Nice find.

Yes. NoodleDoodle.

Hah. Well, I actually meant if there's like a forum post or something offering more details on that. Anyway, cool to hear some news on that front.

I like that painting, I can't believe that no one has bought it yet.

If i had space for it, sure I would buy it

Greg Maxwell asking about privacy and fungibility concerns on reddit/bitcoin here:  https://www.reddit.com/r/Bitcoin/comments/4v28jl/how_have_fungiblity_problems_affected_you_in/

In response to the following question (permalink: https://www.reddit.com/r/Bitcoin/comments/4v28jl/how_have_fungiblity_problems_affected_you_in/d5uv26v)


Answer:


Any Monero devs have any responses, the answer is beyond my technological knowledge?

Lets not forget there are things called compression algorithm's as well if need be. I think the bloat bullshit is being BLOWN way out of proportion!

The short answer here is that this is a linear growth in the block chain in the face of an exponential decline in the resource cost of storing, processing and transmitting data. The combination of linear growth with exponential decline leads to an exponential decline over time in the resource cost of storing, processing and transmitting the Monero blockchain.  Furthermore even in a block chain that is in the clear one would still have the issue of dust outputs that would also grow linearly with time.

Now for the long answer. I will follow up on GingerAle's excellent post which covers very well the history of the mechanical hard drive. First I will quote my signature.
It refers to a punched card storage facility of the United States National Archives and Records Administration in 1959. The Wikipedia article estimates this warehouse held about 4GB of data; however by modern standards of what 4GB of data means this is a gross overestimate. This is because a byte in 1959 punched card and a byte today were very different since the IBM punched card had a very limited character set consisting of the digits 0-9, the capital letters of the alphabet and a handful of special characters. Lower case letters were in practice never implemented in the punched card for example. Furthermore the majority of punched cards were never filled to capacity. With that in mind I doubt that warehouse contained the equivalent of 500 MB of data by todays standards. The IBM punched card was standardized in 1928 and lasted for about 60 years as a "mainstream"  technology. By the mid 1980's it was an obsolete but still somewhat usable technology. Much like the 3.5 in floppy is today. The technology has its roots in the programming of looms in the 18th century and its application to data processing hails back to the 19th century.

The importance of understanding the history of computing technology is that it demonstrated the long term historical trend of an exponential decline in the cost of data processing storage and transmission. In the 20th century we for example saw the transition form the tabulating machines of the 1930's and 1940's to hybrid tabulating machine mainframe computer systems of the 1950's and 1960's followed of course by the mainframe computers of the 1970's and later, mini, micro and personal computers, of the 1970's, 1980's, 1990's and later to the smart phones and tablets of today.

What is most shocking is that while technology has moved by leaps and bounds, business models in most cases have not. The music and movie industries for example are still stuck in the 19th century business models of distributing music on an Edison cylinder by sailing ship, and distributing movies on celluloid film also by sailing ship.  The credit card industry is also stuck in the 1940's business models of overcoming the data processing, storage and transmission costs of the time, by targeting payments made by wealthy consumers purchasing high margin luxury goods and services from high end merchants. This becomes apparent when we see Walmart suing VISA, over merchant fees and no longer accepting VISA in certain stores in Canada. On the other hand we do not see Tiffany & Co. at war with American Express over merchant fees. I wonder why? The critical lesson here is that we must not bake in the current costs of data processing, storage and transmission into the business model of a crypto currency. This is the very serious mistake that many in the Bitcoin community are currently making, and which Monero is avoiding and must continue to avoid.

A reasonable  model for the adoption of a new technology is an S-curve or Sigmoid function. https://en.wikipedia.org/wiki/Sigmoid_function. This starts out as exponential growth, then goes through an inflection point and finally reaches saturation. The adoption of most new technologies have followed this pattern. The issue of "pruning" addresses the linear growth after saturation. to place this into preservative. If we assume an adoption S-curve over say a period of 50 years, this would be comparable to the adoption of payment, credit and debit cards over the last 50 years, then pruning really only addresses a 10x increase in the block chain size over the following 500 years. Is this really an issue? History has indicated otherwise.

Edit: By the way I expect the mechanical hard drive, like the punched card that preceded it, is also after approximately 60 years heading fast towards obsolescence. Solid state drive technology is the likely near term replacement.

3 month bump!

@Smooth, are there any flags that could be set so that the added bloat could be temporary and pruned after a specific time frame or would a HF be required to add one? My mind is weird, I just remembered I never asked this question (I think I took off for a week and forgot about it)!

If I recall correctly my off-chain message proposal could be implemented without a hard fork. It does require nodes to upgrade, or alternately a completely separate program could be used to transmit the "payload" (perhaps in the case of a commerce application this would be a better approach anyway).

As far as Monero stands today, there isn't really any way to put data on the chain that can be completely pruned. Limited pruning of signatures like what Bitcoin currently does (only storage, not addressing bandwidth) can be done with Monero as well (similar to what is implemented in AEON, but it would need to be redone for LMDB). As I understand it, things are reworked a bit in RingCT and more modes of pruning may be possible, similar to Bitcoin's segwit. I have not followed the coding for RingCT closely though, so I don't know the details.

No but it was only a relatively short time in this thread before the above exchange. Shouldn't be hard to find.

The basic idea was to get a hash of the message mined into the blockchain (for example doing a tiny transaction and putting the hash of the message in the payment ID), and paying a fee. The mined hash is then honored by the p2p network to allow relaying that specific message for some limited period like two weeks. This doesn't give a direct incentive to the nodes but at least it minimizes the bloat on the blockchain and prevents cost-free spamming of the p2p.