Copy and Paste from Slack, one of many conversations we have each day.
markpfennig [6:41 PM]
@busoni: hi
busoni [6:41 PM]
Hello
busoni [6:41 PM]
So, the area I'm still fuzzy on is the interaction between marking and the blockchain.
busoni [6:43 PM]
As I understand it so far, when you mark someone, you send some information to a node via the new API
busoni [6:44 PM]
Then what happens--is the information worked into the blockchain?
markpfennig [6:45 PM]
okay, there are two interactions
1. consider the block chain to be the glue between implementations, much as it is currently the glue between poloniex and bittrex, where transactions (trades in your case) are handled off change, but the funds represented can be swapped between different providers
2. every so often a set of markings is hashed up to a merkle root, and that hash used in a blockchain transaction to timestamp it and give some accountability to the off chain transactions
markpfennig [6:46 PM]
then later
3. management of funds in each implementation will optionally be swapped out for key management, allowing marking implementations to send transactions (withdraws and deposts) through a web interface of p2p bitmark nodes
markpfennig [6:47 PM]
so markings are all off chain, as your trades are, but with some accountability on the chain, and the chain as glue between implementations / integrations. In this way we scale massively and are not constrained by block size or chain size, since each integration/implementation handles it's own set of marking data
busoni [6:47 PM]
So what Bitmark offers that BTC doesn't is this accountability?
markpfennig [6:51 PM]
No, that can be done on any chain :grinning:
What Bitmark BTM offers is as follows
1. we're balanced as per the original goals of Bitcoin the paper
2. we don't have any untested tech or greed at the core
3. we have unlimited scope to add what we need, with adoption driving innovation rather than innovation driving adoption
4. we're not targeted at investors or aiming to be some financial market dominated currency, we're more interested in having everybody have 1 BTM each
5. simply, we don't have any constraints using other "coins" would impose, the mark in marking is the mark from btm (0.001 BTM), and whatever marking requires from a cryptocurrency will be implemented as needed
markpfennig [6:53 PM]
anything we do create at crypto level, can be ported back and used on other coins, even our HTTP API for release in Q1 2015 will be included in supernet, and can be patched on to other currencies as needed (edited)
amarha [6:54 PM]
this is a question in the FAQ i'm working on related to this:
amarha [6:54 PM]
Why create a new implementation of Bitcoin, why not just build on top of Bitcoin?
We felt that in order for our project to develop further innovations, and implement other proven open source technologies, it was essential that we had an alternative blockchain that freed us from any political issues that arise regarding the modification of the Bitcoin protocol. Our first relevant example being the intention to modify the Bitmark network nodes to provide a RESTful API. Another future consideration is the likelihood that our Marking technology will benefit from the implementation of side chain technology currently being developed by Adam Back.
amarha [6:55 PM]
bitcoin development might eventually decide to implement marking, but only if it's already successful i would imagine
markpfennig [6:55 PM]
additionally we don't aim to compete with alt coins, more standardize what they're doing. If our user base later demands something like 'anon' at the core, then we'll find the best solution, ensure it's matured and add it to the core. We're not trying to innovate to compete with alts.
busoni [6:55 PM]
What exactly do you mean by "implement marking"
markpfennig [6:56 PM]
Marking is a web level protocol
busoni [6:56 PM]
It sounds like it's pretty separate from the blockchain to me, though
markpfennig [6:56 PM]
it is!
markpfennig [6:56 PM]
@busoni: have you read
https://github.com/project-bitmark/marking/wiki ?
GitHub
project-bitmark/marking
marking - Future home of Marking
amarha [6:56 PM]
eventually the each node will be involved, also the issue of mass writing data to the blockchain as seen in the issue with counterparty and many of the core bitcoin devs
busoni [6:57 PM]
So what would it mean for Bitcoin to "implement marking" if it's separate from the blockchain?
amarha [6:58 PM]
integrate the changes to the nodes i think would probably be necessary eventually, but mark knows better than me
markpfennig [6:58 PM]
It would be the inverse, to allow Marking to use bitcoin, which will be enabled later on once it is mature, and once all requirements of a coin daemon have been defined and implemented.
Likely it will require a patch to the core to include the HTTP API we create, and a couple of lines to change in marking implementations
markpfennig [6:59 PM]
from day 1 we have included amount+unit so it can be used with any crypto, or perhaps even fiat, or apples and oranges and grains of rice (edited)
busoni [6:59 PM]
And this accountability on the blockchain -- does that require a special modification?
markpfennig [7:00 PM]
to be determined, I'll be able to answer that succinctly in the next few weeks, it may well require a modification to transactions, but hopefully not if we can get away with it
amarha [7:00 PM]
so far every effort has been made as far as i know to keep bitmark 100% backwards compatible with bitcoin
pabloangello [7:01 PM]
https://pbs.twimg.com/media/ByDff6wCcAAgrKd.jpg:large (20KB)
pabloangello [7:01 PM]
lol nice
nocoins [7:01 PM]
Not a fan :confused:
pabloangello [7:01 PM]
with this dots it looks like Panda :stuck_out_tongue_winking_eye:
markpfennig [7:01 PM]
@busoni: interestingly the approach may be applicable to you at poloniex, it may provide a way for you to expose chained trades together and have them timestamped in the chain, it's kind of like a sidechain which needs no mining
amarha [7:02 PM]
couple other FAQ's in the works:
amarha [7:02 PM]
What is Marking?
Marking is a web-scale platform for the future of the web(the semantic web or web 3.0). It enables the distribution of marks(one mark is also 0.001 BTM) in a frictionless and viral manner. It’s also reputation+trust system backed by cryptocurrency. In it’s simplest form it can be viewed as spendable karma that people share for creating and sharing content, or as a microtransaction platform. As the system evolves, more complex trust systems can be built on top of marking. One example is our plan to integrate Martti Malmi’s(sirius on bitcointalk) Identfi project in to marking. The marking data is extensible, and the possibilities of how marking can be used are unlimited.
Why marking?
Project Bitmark started as a project that was focused on making cryptocurrency accessible and usable as a daily use currency. Marking is our first major initiative to make that happen. (edited)
markpfennig [7:02 PM]
s/out/our
amarha [7:03 PM]
it's still early alpha :stuck_out_tongue:
amarha [7:03 PM]
What can be marked?
Anything. Although if you wanted to mark a location, you might mark it with the minimum amount possible, as a way to register your mark on the network so that data can be used by the many services that might make use of your location mark. Whereas in a p2p marking transaction in certain situations, such as marking someones music, or art, the value of your marking(or the value you're receiving from being marked) could be much higher.
What do you mean by marking is reputation?
All the marks that a person, a place, an object, or an abstract concept even receives gets recorded in the local database of that marking integration. From there each and every marking that takes place is hashed as a Merkle Tree and included in the Bitmark blockchain. So the marking data can be verified, shared, consulted, analysed or downloaded forever.
amarha [7:04 PM]
roughly correct Mark?
busoni [7:04 PM]
Does this depend on the honesty of the sites/service implementing marking?
busoni [7:04 PM]
Say one site lets you mark a location... then another site lets you mark the same location.
busoni [7:05 PM]
Will the location have two marks, somehow? Will they be combined?
markpfennig [7:05 PM]
wonderful questions :smile:
busoni [7:05 PM]
And if so, what's to prevent one service from abusing this?
markpfennig [7:07 PM]
honesty: yes and no, it's a reputation system and we are also integrating identifi to give trust metrics. So the flow of marks in and out of a service/implementation can serve as a measure as to it's performance, and the trust statements about it can provide information as to whether it's to be trusted or not.
our goal is not to ensure that people are good or bad, but to provide enough data (a graph of reputation+trust) so that people can build tools on top to address these issues. Specialists in thos areas will do better than any attempt we make.
busoni [7:08 PM]
So Bitmark itself won't comment on reputation
busoni [7:08 PM]
Just provide the data for implementations to deduce reputation
markpfennig [7:08 PM]
definitely not, each person will as needed, we're decentralized
amarha [7:09 PM]
the tools should be there for people to analyze the data and build systems on top of it to help determine trust
busoni [7:09 PM]
And all marking data will make it to the blockchain eventually?
markpfennig [7:09 PM]
no
markpfennig [7:10 PM]
never, it's on top
amarha [7:10 PM]
at a raw level, with out identifi or any other data analysis tools, you can look at it like spendable karma(like reddit)
amarha [7:10 PM]
but the potential for it to evolve in to much more is there
markpfennig [7:10 PM]
regarding two different implementations providing the ability to mark the same things:
this is our goal, it's free speech and how the web works. Each of the services may provide different information or purposes of marking, for example on the web you may look up about "France" in different contexts, a general one at wikipedia, a more specific one on a govt site, and a more social one on trip advisor
Marking takes the same approach, so one service may provide markings related to say an event which happened at the location, and another to some geographic feature, and another to some photo taken at the place at some time, and so forth (edited)
amarha [7:12 PM]
we're also going to be creating our own marking implementation at getmarked.org which should provide tools and data, and also aggregate content online and reward those who create it and share it
markpfennig [7:12 PM]
each implementation can be seen as specializing for some purpose. Also "markings" the data structure, is extensible, so further properties for some purpose can be added, which allows people to say what they like about anything, some apps will show some types of data, and others other kinds, but the core set of properties remains common, the stuff that makes it function (interoperability) (edited)
amarha [7:14 PM]
one of the major goals is to get regular people using cryptocurrency, without even realizing they are. then when they decide they want to redeem some of the marks they've earned, they can begin to learn about the blockchain behind the scenes.
busoni [7:14 PM]
Just as confused as ever about the involvement of the blockchain...
amarha [7:14 PM]
i meant them sending it to hold them theirself
busoni [7:14 PM]
It seems sometimes you're saying that the data will be included in the blockchain, other times you're saying it has no connection with it
amarha [7:14 PM]
when they're using marking locally on a webite, it's not
markpfennig [7:14 PM]
okay, I'll clarify simply
amarha [7:15 PM]
but the blockchain is what gives marks their value is what i meant there by the blockchain behind the scenes
markpfennig [7:16 PM]
the data is outwith the blockchain, each implementation handles it's own set of markings
each marking is a balance transfer from one user to another within that implementation
the implementation manages balances
(all the same as poloniex does with trades)
the blockchain is for sending marks out to storage or back, and between services/implementations
as a secondary usages, sets of markings are hashed up in to a tree and the hash of that structure is timestamped in the block chain, by sending a tx to the hash.
markpfennig [7:17 PM]
the block chain isn't needed for marking, if it was it wouldn't scale
markpfennig [7:17 PM]
but it is needed to keep the thing decentralized and allow marks to flow around between users and services and storage and back
markpfennig [7:18 PM]
i.e. I can +10 you here and you can `withdraw <btm-address> 10` and get it in your wallet or account at poloniex
markpfennig [7:18 PM]
think of it...
markpfennig [7:19 PM]
as being able to withdraw your likes at facebook, your karma on reddit, your favs on facebook,. and give some of them to your mum, some to a mate via email, and use some to buy a coffee..
busoni [7:19 PM]
Okay
busoni [7:19 PM]
That's the version where it has nothing to do with the blockchain
busoni [7:19 PM]
Now what's this business about sidechain technology?
markpfennig [7:20 PM]
hmmm...
amarha [7:20 PM]
that was just a reasoning to use our own chain
amarha [7:20 PM]
potentially wanting to do something like that
busoni [7:20 PM]
So the sidechain thing isn't set in stone yet
amarha [7:20 PM]
in the future, to have control
markpfennig [7:20 PM]
@busoni: is it fair to say you're looking for some novel technology at blockchain level to warrant the existence of BTM?
busoni [7:20 PM]
No
busoni [7:20 PM]
I'm just trying to understand it fully
markpfennig [7:20 PM]
great :smile:
busoni [7:21 PM]
Part of that is knowing about any blockchain-level changes
amarha [7:21 PM]
it all depends on how the development goes on sidechains, if they are useful for marking, then i think mark will want to implement them
amarha [7:21 PM]
they are just one of the potential techs that potentially matter
markpfennig [7:22 PM]
okay for the sidechain thing, it's an alternative approach.
Blockchain is a distributed time-stamping server
we need to timestamp markings, and give a link back to that set of markings
so we figured that we can hash up markings in to a structure, calculate a hash, and send a mark to it on the chain, with a message that points back to that set of transactions on the web (or wherever)
markpfennig [7:22 PM]
i.e. use the blockchain, the timestamping tech, as a timestamping tech
busoni [7:22 PM]
What exactly are we timestamping
markpfennig [7:23 PM]
sets of markings
busoni [7:23 PM]
A transaction, like a withdrawal of marks?
markpfennig [7:23 PM]
let's convert this to trades at poloniex to describe it
markpfennig [7:24 PM]
if every hour you took all trades, and published them as a list in sequence, with say the hash from trade-1 as a property of trade-2, and the hash of trade-2 in trade-3
then you'd have a list in a provable set order, that trades happened in this order
then if you took the set of trades for that hour, and hashed them all up to get the sha256 sum, any modification to the data would produce a different hash sum, so you'd know it had been tampered with
markpfennig [7:25 PM]
now you submit that hash to the blockchain, as the address for a transaction, and send a mark to it
busoni [7:25 PM]
I'll be back soon... apparently there's a small Polo crisis.
markpfennig [7:25 PM]
np
amarha [7:25 PM]
thanks for coming by :grinning:
markpfennig [7:25 PM]
now looking back from the other direction, you can see the tx, follow through to the data, and verify that it hasn't been tampered with, that no trades are changed or out of order
markpfennig [7:27 PM]
if you can grok how that would provide the accountability to your exchange that users may want, then you can see how the same approach for markings (or any data) can be used to provide accountability and tamper proof data (or at least data where you can prove it has been tampered with), trust+accountability
markpfennig [7:28 PM]
@busoni: the above finishes it, unless you have more questions - GL with polo issue
melvincarvalho [7:39 PM]
nice discussion, good questions from @busoni
markpfennig [7:40 PM]
agree completely :grinning: this is why poloniex was out first choice, he's interested in the tech and projects not the money
markpfennig [7:40 PM]
(or more than the money)
melvincarvalho [7:43 PM]
markpfennig: any particular reason you'd want to hash the merkle root in the (i assume btm) block chain, in satoshi's paper he talked about publishing widely, e.g. on usenet, but why would we not use github for example?
busoni [7:45 PM]
Okay, that finally makes sense
markpfennig [7:45 PM]
@melvincarvalho: just to check we're aligned in thought. It would be submitting say the hash of a release on github to the blockchain for timestamping, so commit <x> can always be timestamped as happening at the time it was, and hasn't been tampered with
busoni [7:46 PM]
It sounds like you're still working a lot of specifics out
markpfennig [7:47 PM]
@busoni: yes very much so, it's work in progress and even once released we'll be working out bits
melvincarvalho [7:47 PM]
im new (ish) to the project so just learning still ... :grinning: (edited)
markpfennig [7:47 PM]
the system has so many applications it's hard to account for them all
markpfennig [7:48 PM]
conversations and q&a sessions like these help a great deal (edited)
busoni [7:48 PM]
It's a little hard to wrap one's head around when we're so used to blockchain-level changes with alts
busoni [7:48 PM]
This is bigger than that
melvincarvalho [7:48 PM]
i understand the role of the block chain as a glue between federated instances, this is great. I understand the role as a coinbase to create monetary units, also great. I'm still slightly unclear on what further use it has in relation to off chain tx/uxto/ledgers ... (edited)
markpfennig [7:49 PM]
@busoni: yes :grinning: and it's quite nice, for the first time in years many of the crypto people here can talk about things with their friends and family, you can discuss marking wit your wife/mates/kids and they get it. It's the crypto people that have to unlearn
markpfennig [7:50 PM]
@melvincarvalho: it doesn't it's an optional part. Hash up some data, get the sum, send a mark to the address corresponding to that hash and it's timestamped, can be used for anything not just marking
ace [7:50 PM]
Much bigger.
melvincarvalho [7:50 PM]
got it
melvincarvalho [7:51 PM]
this can already be done
melvincarvalho [7:51 PM]
and is done in the btc block chain
markpfennig [7:51 PM]
yes
busoni [7:51 PM]
It's not actually that complicated, it's just thinking in a fundamentally different way that makes it so hard to grasp at first -- as you said, for crypto people.
markpfennig [7:52 PM]
agreed
melvincarvalho [7:52 PM]
true
ace [7:52 PM]
It's also because crypto has taught ppl to doubt first
ace [7:52 PM]
believe later
busoni [7:52 PM]
We need to think about this more, but I think implementing marking on Poloniex is likely something we want to do.
busoni [7:52 PM]
Exactly how is what we need to think about
melvincarvalho [7:53 PM]
crypto has taught people the zero trust paradigm, but zero trust can exist in parallel with micro trust, this is what happens on exchanges of course
busoni [7:53 PM]
In the trollbox is an obvious start, but mobydick was talking about marking markets
markpfennig [7:53 PM]
we can help there, you may even be able to use the getmarked api for it.. but it may make more sense to just do balance transfers between users, you already have 50%+ of the functionality
markpfennig [7:53 PM]
marking markets sounds cool
markpfennig [7:54 PM]
other things, like groups of traders all marking a proposed trade, if enough of them agree trade is executed - like group managed funds
markpfennig [7:54 PM]
the marks themselves could be the funds!
busoni [7:55 PM]
Uy... this is where it gets complicated. There are just SO many applications for marking.
markpfennig [7:55 PM]
lol yes, infinite, in real life too
ace [7:55 PM]
would the amount of marks someone has be displayed?
busoni [7:55 PM]
That would be an idea for the trollbox
busoni [7:55 PM]
A reputation system for users
markpfennig [7:55 PM]
ace, depends, it can be layered on, or default amounts like 1m always
ace [7:55 PM]
I feel as if it's Bitcoin is to fiat as Marking is to indexing
ace [7:55 PM]
Its solution
ace [7:56 PM]
Marking has the potential to index all the data in the world
ace [7:56 PM]
in some form
markpfennig [7:57 PM]
still grasping it myself, and I have conversations like this daily..
markpfennig [7:57 PM]
two from yesterday...
markpfennig [7:57 PM]
1. free education where the students mark the teacher
markpfennig [7:58 PM]
2. marking data from different bio/med datasets, providing bounties for finding new medical breakthroughs by incentivising the merging of data from different research units
markpfennig [7:58 PM]
in every convo, it seems to flip the way humans do things on it's head
wonderloops [7:59 PM]
its the beauty of it, people will use it as they see fit and build their own applications
markpfennig [7:59 PM]
yup, we don't need to, others will have far better ideas than us
markpfennig [7:59 PM]
will be fun to watch :grinning:
markpfennig [8:00 PM]
already is..
wonderloops [8:00 PM]
shouldnt be our goal to try and develop a 100 idea's to proof its possibl
markpfennig [8:00 PM]
the whole of project bitmark is about rolling out the ideas in some way and moving to the next, or integrating with whatever software we can find, to enable others to do their ideas
markpfennig [8:01 PM]
we don't really have much work to do other than that once marking is stable, bitmark/btm already is
markpfennig [8:01 PM]
just adoption :grinning:
markpfennig [8:01 PM]
'tis what we mean by "focused on adoption"
ace [8:02 PM]
+ 10
ace [8:02 PM]
oh forgot
markpfennig [8:02 PM]
okay, hungry children I must head off, I'll catch up before I start coding - it's about time we made these discussions reality
ace [8:02 PM]
only in general
ace [8:02 PM]
Take care mark!
markpfennig [8:03 PM]
the same to you all, and thank you so much for all your support and time. For a while back in July I thought we might never reach this point, it was getting tough.
ace [8:03 PM]
Hard works pays off Mark
busoni [8:03 PM]
Thanks for taking the time to explain things
markpfennig [8:03 PM]
Who composes the dev team?
Project Bitmark is a community effort, currently there are roughly 50 people in various roles and having different levels of commitment; this number grows daily. Additionally we have outreach to various projects within and more importantly outwith the crypto space, and businesses which are being created around what we're doing. Full time we have Victor (Medic) our project manager, Aureus (EsteNuno) our community manager, and two full time developers at present Melvin Carvalho and myself, Mark. It is a shame to only mention a few names when so many have committed and contributed.
ace [8:04 PM]
your results are well deserved
markpfennig [8:04 PM]
our*
ace [8:04 PM]
Haha fair enough
markpfennig [8:04 PM]
:wink: back in a bit
ace [8:04 PM]
Enjoy!
melvincarvalho [8:05 PM]
i think satoshi actually describes this in his paper page 1
melvincarvalho [8:05 PM]
"Commerce on the Internet has come to rely almost exclusively on financial institutions serving as trusted third parties to process electronic payments. While the system works well enough for most transactions, it still suffers from the inherent weaknesses of the trust based model.:
melvincarvalho [8:06 PM]
Line 1 in fact
melvincarvalho [8:06 PM]
here's the key "the system works well enough for most transactions"
melvincarvalho [8:07 PM]
The block chain "allowing any two willing parties to transact directly with each other without the need for a trusted third party"
melvincarvalho [8:07 PM]
again the key "willing parties"
markpfennig [8:07 PM]
yes! the systems are not mutually exclusive - we have to work with what already exists, and what will exist in the future, not try to replace it all
melvincarvalho [8:07 PM]
user gets to choose
melvincarvalho [8:07 PM]
not the developer
markpfennig [8:07 PM]
+10
