Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - maqifrnswa

Pages: 1 ... 9 10 11 12 13 14 15 [16] 17 18 19 20 21 22 23 ... 45
226
Technical Support / Graphene and secp256k1-zkp
« on: June 23, 2015, 09:39:00 pm »
I see that the secp256k1-zkp library has configuration options hard coded instead of using the ./configure script:
https://github.com/cryptonomex/fc/blob/master/CMakeLists.txt
https://github.com/ElementsProject/secp256k1-zkp

That is causes build errors on some systems:
In file included from /home/pi/graphene/graphene/libraries/fc/vendor/secp256k1-zkp/src/secp256k1.c:11:0:
/home/pi/graphene/graphene/libraries/fc/vendor/secp256k1-zkp/src/util.h:102:37: error: ‘__int128’ is not supported for this target
 SECP256K1_GNUC_EXT typedef unsigned __int128 uint128_t;

since not all architectures have __int128

just running ./autogen.sh and ./configure then make works in that folder, but the fc cmake build does not work

I know it's early and it's probably just set up for testing at this point, but I wanted to give a heads up.

227
I'm also getting this error early on:
In file included from /home/pi/graphene/graphene/libraries/fc/vendor/secp256k1-zkp/src/secp256k1.c:11:0:
/home/pi/graphene/graphene/libraries/fc/vendor/secp256k1-zkp/src/util.h:102:37: error: ‘__int128’ is not supported for this target
 SECP256K1_GNUC_EXT typedef unsigned __int128 uint128_t;

which, according to:
http://raspberrypi.stackexchange.com/questions/3919/using-128bit-integers-in-gcc
__int128 isn't defined on raspberry pi

228
Technical Support / Re: 2.0 local server via a raspberry pi?
« on: June 23, 2015, 05:14:15 pm »
I might get this done sooner than I thought...

I've been piecing together how to emulate raspbian without actually needing hardware, and without using qemu to emulate a whole system (which is so slow) and wrote:
https://wiki.debian.org/RaspberryPi/qemu-user-static

The image I'm working on will be a full jessie rasbian image you can just flash to your sd card. I will have all the commands for the bitshares 2.0 cli wallet and witness node. We'll need to keep an eye on zhangweis' problem too:
https://bitsharestalk.org/index.php/topic,14917.msg193170.html#msg193170

229
Technical Support / Re: 2.0 local server via a raspberry pi?
« on: June 22, 2015, 08:44:17 pm »
Sounds like a fun project - I can get one too (just don't have one at the moment). I can try it out and cass can help with testing.

I think this is what we have to work with for now:
https://github.com/cryptonomex/graphene

Also, I mean RPI 2, not "B" in my previous posts... And we could just base it off of raspbian too, that's easy also and would get better coverage right away... If I get a chance this week, I'll see if I can whip something up for pi owners using raspbian to test.

230
Technical Support / Re: 2.0 local server via a raspberry pi?
« on: June 22, 2015, 06:56:15 pm »
You can also sell pre-loaded SD cards so people can plug them into RPis. If you want maximum distribution, creating a simple image that can be downloaded and put on an SD card is trivial. Especially since debian armhf natively supports RPI Bs. Just install debian armhf on a RPI B, build bitshares, set up ssh, make some systemd services to start/restart bitshares - and then wrap up that image for distribution. Someone downloads the image, copies it to their SD - sticks it in, and it's a headless, stand-alone server that starts once you plug it in.

BONUS: the RPI B image will probably work with many others boards with no changes, including cubie board, some of those omex (sp?) boards, etc.

sounds great. can you do it? :D

yes :-) It's only a little more work than running the PPA. I'd need a PI B, I only have an A. (it could work on the A too, but the newer ARM chip on the B is easier to use with existing pre-compiled libraries).

EDIT: keeping it up to date would be harder... Sure, I install my image - but then it's stuck at that version. It might be easier to just set up our own debian armhf "ppa" and tell people to just install that package - or to have our pre-built binaries just go and fetch the up-to-date releases from that package. We can't get it into Debian main/Ubuntu universe because of the "you must use this with bitshares" clause, but "non-free" would be acceptable. In that case you just tell the Pi to grab the released package from the repositories.

231
Technical Support / Re: 2.0 local server via a raspberry pi?
« on: June 22, 2015, 05:50:28 pm »
You can also sell pre-loaded SD cards so people can plug them into RPis. If you want maximum distribution, creating a simple image that can be downloaded and put on an SD card is trivial. Especially since debian armhf natively supports RPI Bs. Just install debian armhf on a RPI B, build bitshares, set up ssh, make some systemd services to start/restart bitshares - and then wrap up that image for distribution. Someone downloads the image, copies it to their SD - sticks it in, and it's a headless, stand-alone server that starts once you plug it in.

BONUS: the RPI B image will probably work with many others boards with no changes, including cubie board, some of those omex (sp?) boards, etc.

232
And, for the third time: As long as narwal actually creates a market for them and works to develop the infrastructure that can support it, good for him/her.

This isn't about "we don't like what he did let's take away his property" (at least to me). In fact, I supported narwhal in his efforts to monetize the names. I have no problems with squatters, as they perform a market function and are entrepreneurs that took a risk. I do have a problem with systems that enable perpetual squatting. That is a bad game theory/economic model. It's like margin trading with no interest or calls. If names are a resources, I don't want the blockchain to give the resource away forever. See puppies post on this topic:
https://bitsharestalk.org/index.php/topic,17016.msg218242.html#msg218242


Now, for the benefit of the community, a lesson in how to have a fruitful discussion
Quote
You tell me which argument is the straw man:

1. Now that we've finally caught the diabolical Molecatcher, what should the community do to punish him (mwoo hoo woo oo ah ah ah ah!)?

2. What should the account name registration fees be?

Neither are arguments, neither are strawmen. For an understanding of what an argument is, see;
https://www.youtube.com/watch?v=kQFKtI6gn9Y
which - although funny, drives home the point of how an argument (point of discussion) differs from an argument (discourse between opposing positions). A straw man argument is an intentionally weak point of discussion set up by one party to make the other look weak. For example:
"The community lacked foresight in their fee structure, narwhal took advantage of the fee rules, the community feels like they got ripped off, therefore the community wants to whine and take away people's property" is a strawman argument that you can take down with, "As bitshares we all agree we must protect property, so everyone that wants to take away property rights is against bitshares" And thus avoid the arguments of, "all rules are defined by the blockchain, the genesis block of the new blockchain has not been written yet, therefore there are no rules yet for the new blockchain" or the argument "while there is market value to name speculation, there is no market value to indefinite name speculation or removal of names from the namespace, therefore indefinite name speculation should be considered for possible removal"

Quote
Further discussing what to do with Mr. Molecatcher is indeed a "strawman argument" in comparison to the pertinent discussion on consensus building to secure the integrity of BitShares and your childerens' future.  I have been hammering you on this for over a year now, and it is exhausting.  But that is what this community loves to do...

Again, that is not a strawman argument (nor an argument at all) - you are confusing the two definitions of argument:
https://en.wikipedia.org/?title=Straw_man

233
General Discussion / Re: New accounts last 24h:
« on: June 19, 2015, 11:38:50 am »
This is a democracy.  It is a democracy with hard coded rules that we can vote to change.  If all the Narwhals go extinct, and all the narwhal names go unused we can deal with that in the future.  You know, for when we really need the name molecatcher.  I see no need to mess with the narwhals expected business model at this point.

It brings up a very important point though.  Registered names are by definition a scarce resource as they can only be used by one person at a time.  When we register a name we are effectively staking a claim to them.  Common law for claims of previously unowned property required both enclosure, and development.  Otherwise the property went back to its original unowned state.  The blockchain securely encloses this scarce resource for us, but does not enforce development.  What means of enforcing development would the community find palatable?

One model that springs to mind is that of a black hole.  They are supposedly constantly losing mass due to hawking radiation.  For a more massive black hole that is not really an issue, as it will take billions of years to waste away.  For a micro black hole this become an issue though.  If it evaporates too much it will vanish.  Could we do something similar with names.  Require a certain balance, to be parked on names that is burned as a fee at a certain time increment  Perhaps just enforce a certain minimum fee burned within a time frame.  If an account name does not have the fee it is purged from the blockchain.  This would dramatically reduce squatting.

Perhaps the answer is to split account names from domain names.  Require that domain names are linked to actual ip addresses.  This could be as simple as requiring domain name owner keys to broadcast signed transactions to the network every month or so.  Adding a small amount of work and cost to maintaining a name.  We could try to go farther and require the mapped IP to be provably under the control of the owner key of the domain.  Perhaps embedding this signed message into the page itself.  Then we would want to limit the number of domains that could be assigned to a single IP. 

While there is no reason to believe that the large aquatic animal is lying to us, I think we should disabuse ourselves of the notion that if we lay claim to something we can own it forever without putting forth any further effort.  If all it takes is laying a claim, then I own the Moon, Mars, Venus, Jupiter and all its moons.  In fact I own the Sun, and I need all of you to stop taking advantage of my property.

 +5% this is my concern, well articulated. Has nothing to do with narwhal, he can keep names or not, whatever, but this is the problem I see with the system thanks to narwhal illustrating it.

234
General Discussion / Re: New accounts last 24h:
« on: June 18, 2015, 06:15:49 pm »
My concern is that for only $100 someone can reserve all those names then go AWOL, thus losing those names from bitshares forever. As long as narwal actually creates a market for them and works to develop the infrastructure that can support it, good for him/her

Raise the price.  Drop a hard fork tomorrow.  This is still a centrally controlled coin.

Problem solved........

What if you go AWOL?  We all die.  Let's shut down the whole project because death is inevitable?

So what.  Nobody will ever get to use the name molecatcher.  I say, let him keep going.  When he moves on to the Spanish dictionary, he will have to buy more BitShares on the open market.  Isn't that the point?  He hasn't even begun forming complete sentences yet.

Oh wait...sorry, I forgot...We hate making profits around here...

BitShares T - "oh shit! we're making profits!  SHUT HER DOWN!"

What is this, Bizarro Business 666?

I think we must have skipped Business 101 the day they taught:

"accepting profits"

Here's the Cliff's Notes:  (you're supposed to choose to keep them)

If I go AWOL, I do understand the grief and pain it will undoubtedly inflict on the psyche of all internet users, but that is besides the point and a burden I live with every day.

I think you're playing with strawmen a bit much. The question isn't just business, but technical - both must be balanced. It may be possible that the better business decision is to protect the technology while it is in specification discussions even if it means a single user loses money on a bad bet (the bet was that the specification wouldn't change before implementation, which is counting your chickens before they are hatched)

I'm not saying shut anything down, just an honest academic question. Which blockchain/business is better:
one where you can register any name at market value, but that "screwed" one person while the protocol was still in discussion and not implemented yet
OR
one where it is forever impossible to register ~40-80k names because they are lost forever, but kept it's principles and let someone lose those names. And this includes molecatcher, which would be a catchy name for an exterminator service.

At the same time, and I repeat:
As long as narwal actually creates a market for them and works to develop the infrastructure that can support it, good for him/her.

EDIT: speaking of business 101... It's not uncommon for performance metrics when granted a license. Narwal has purchased a license to all those names; I wouldn't be opposed to a performance clause on the usage of those names. Imagine if someone registered citibank for 0.10BTS and then never logs in again. That would significantly hurt the blockchain and bitshares. About "accepting profits," if I was to go to citibank and give them $100, do you think they will give me their building in Manhattan? I guess they skipped business 101 as well.

235
General Discussion / Re: New accounts last 24h:
« on: June 18, 2015, 03:32:47 pm »
My concern is that for only $100 someone can reserve all those names then go AWOL, thus losing those names from bitshares forever. As long as narwal actually creates a market for them and works to develop the infrastructure that can support it, good for him/her

236
General Discussion / Re: The Matrix Trilogy Decoded - Mindblow
« on: June 17, 2015, 04:43:42 pm »

Let's me decode this for you all...

You will either be on the blockchain know it, or you will be on the blockchain don't know it.

237
Stakeholder Proposals / Re: Short Order Refactoring
« on: June 16, 2015, 09:09:43 pm »
This could be the mechanism that would allow arbitrage towards the feed - it could be the missing link!

When BTS:BTA is greater than the feed, I can generate BTA for cheaper than I can buy them on the market, then sell them on the market for profit, which returns the price to the feed! You are rewarded for "fixing" the feed when BTA is under-supplied!

238
Technical Support / Re: BitShares 2.0, Graphene & Cryptonomex
« on: June 16, 2015, 05:51:09 pm »
and I thought it was a good deal before... And it's much easier when you use established licenses like CC0 and BSD!

239
General Discussion / Re: Cryptonomex? WTF is this?
« on: June 15, 2015, 01:29:03 pm »
Here is the deal, "BitShares" as not a legal entity and cannot "own" IP.    I personally do not recognize IP, but most people do including some of those who have contributed to BTS source code.
The public domain is the legal entity that owns BitShares IP. It's a very real legal entity and you are infringing on its rights by trying to alienate its enjoyment of the intellectual properties it owns.

Public domain is not a legal entity, nor is it even a license granted to anyone. It is just the relinquishing of copyright. It is impossible to infringe on the rights of the public domain. You can't prosecute anyone for violating anything because there is nothing to violate and no one with standing to enforce. It is perfectly legal to take public domain software and charge people $100 to download it from you.

Quote
Beside your stance regarding BitShares legal status is inconsistent and changes depending on the point you are trying to make. It's not a legal entity when it's about stripping it from it's intellectual property. But when you are pitching BitShares 2.0 BitShares is a well defined entity that can be the legal recipient of a software license with all the guarantees that this implies. That doesn't add up. If BitShares is not a legal entity, there is no way that you can make a software licence that will legally entitle BitShares to unlimited use of Graphene since BitShares is something that can't quite be define legally. That means that the exclusive license is a castle of cards that won't stand in court.

You are right here, this needs to be cleaned up. Graphene should NOT be licensed to BitShares since BitShares doesn't exist. instead, it should be licensed to the end users/future developers/anyone that downloads the code who are free to modify, distribute, and use with no restriction provided it is used with the longest bitshares blockchain. That is fine, and will stand up in court.

Quote
So let's be clear here and stop sugar coating.

BitShares not being a legal entity doesn't own the intellectual property but that doesn't mean that the core developers own it instead. As per the original license under which BitShares was released, the property rights to the work belong to the public domain. Authors retain only the moral rights.

Yes, that is correct - but I don't see what the problem is. No one is claiming ownsership of the original bitshares code, just the as-of-yet unreleased modifications. That is perfectly legal (provided the funding came from independent sources and development)

Quote
BitShares not being a legal entity cannot be protected by any software license whatsoever, so all the reassuring guarantees of exclusive and illimited licensing are hollow promises that can be revoked on a whim by Cryptonomex and won't stand in court.

You are correct, the license is not TO bitshares but given to the users and future developers of bitshares. Those are not hollow promises and those do stand up in court.


Regarding ownership of intellectual property, there are three cases out of which two are straightforward:
- All the code and ideas that are implemented in the code of the current BitShares project belong to the public domain as per the terms of the Unlicence under which it has been released. There is no way back on that, so Cryptonomex can't claim any particular rights on that code and set of ideas, and can't patent anything that makes the current BitShares protocol what it is since BitShares stands as prior art. The public keeps all rights of reuse of the existing code and owes nothing to Cryptonomex regardless of the claims they may make.
Yes
Quote
- All the code and ideas that are developped independently by Cryptonomex using its own funds and its own resources belongs to Cryptonomex and they may use and license it as they see fit.
Yes
Quote
- For all the code and ideas that hasn't been put in the public domain but has been developed prior to the incorporation of Cryptonomex or by individuals not affiliated to Cryptonomex at that time, and in particular when it could be shown that all funds meant to fund the development of BitShares hadn't been used yet, the ownership isn' clear. It may belong to its respective authors or to BitShares and therefore the public domain depending on exact circumstances. Although there really isn't much legal guidance about that case since BitShares isn't a legal entity and hasn't signed non-compete and disclosure of conflict of interests agreements, it would still be easy to show that appropriation by Cryptonomex of code which development was funded by the public for the purpose of releae in the  public domain would be unethical, so the outcome of an eventual judgement on the matter is anything but clear.
(highlighting by me) That is where the problem can exist. Usually companies set up systems to not contaminate development of one funding source with another. I'm assuming that the reason cryptonomex was set up was to isolate I3's work (public domain) from the new work (cryptonomex)

240
General Discussion / Re: Cryptonomex? WTF is this?
« on: June 15, 2015, 01:20:32 am »
Oh man you are such a breath of fresh air every single time you speak.

BM and the devs are trying to navigate a crazy space of the existing legal structures/regulations and BitShares, which figuratively breaks the mold. This takes creative licensing, IP, corporate structures - all things that are against anarcho-capitalistic motivations that drives many people to cryptocurrency.

Honestly, if the devs didn't do this, bitshares would be more likely to fail to reach its goals.

This is the economic/business development equivalent of using copyrights to prevent IP restrictions
https://en.wikipedia.org/?title=Copyleft

It's still an experiment, the ideal mode is unknown - but as long as all the actors are true believers, they can work together to find a local (if not global) optimum.

Pages: 1 ... 9 10 11 12 13 14 15 [16] 17 18 19 20 21 22 23 ... 45