New Era for Lisk

New Era for Lisk

1 comment

I played today on a fresh install of a testnet lisk node on a raspberry 4 last generation, It's a long and complex game as i like it !
For this 4GB machine running Raspbian Bullseye on ARM only one method to install lisk : sources ! So i did it
Recipe :
Take a Raspberry, at least 1GB memory and at least 16GB hard storage ( SD card or external USB storage)
Install Raspbian last version
Follow Instructions, mix with instructions on github because both are not enought but with both you may manage !
Nodejs 12.14.1
I did install Postgresql 11 instead of 10, without issue so far.
If your Rasberry is on your home network, think to add it in your DMZ.
When first start, it will fail until you discover you must set the db password inside your config.json.
Loading the first snapshot blockchain.db.gz is long, very long. very very long !

But finally, it is going well, and forging. Obviously it is only a game i won't replicate on mainnet !

No comments

I am a afraid I remain the only one to fight against this calamitous state on lisk dpos. A cartel invaded it years ago.
They won, obviously. On official lisk medias, complaining against the cartel is forbidden. It is as in a bad dream: Bads won, Bads became goods. After a battle, the winner is always the good guy.
Discord official channel is now "cartel compliant", as Reddit.
I am afraid I am the last remainder. With my small power, my poor English.
Other delegates won't complain, too happy to collect rewards mainly for themselves.
Standby delegates, even "Heroes", won't complain. Scared to be banned by the cartel's votes.
A new dpos is not an option, it is an urgency

No comments

I am working in automotive industry for more than twenty years. From IT systems and networks, i have been involved since in many projects around ERP for automotive. And then i've been working on many areas within this industry : purchases, sales, production, planning, MRP, financials, accounting, logistic, ... and EDI.
EDI for Electronic Data Interchange is not even essential for many industries, it is obviously strategic for automotive industries. If EDI systems stops, very quickly everything will die. My concern here is only about EDI for sales, purchases and invoicing messages, hence :

  • DELFOR : where customers send their orders, firm and/or forecasts
  • DELJIT : where customers send their orders for JIT cards. A JIt is a logistical entity, encapsuling a product, a quantity, a schedule, and eventually many others things besides.
  • DESADV : where a tier, customer or supplier, is telling what he has sent, or what he has taken in a particular warehouse
  • INVRPT : where a tier is telling about a particular logistic operation
  • INVOIC : where a tier is sending an invoice

Other messages could exists, but with these five above, we've got essentials.

When sending a message B2B from X to Y, traditionnal path is :

  • ERP X is building it's own message
  • Local EDI station X is translating the message to a normalised syntax (DESADV 96A, VDA, ANSI 12, or whatever !)
  • The EDI msg is sent in particular communication systems, or any encrypted subnetwork via internet. Many companies are working in this area.
  • EDI station of receiver Y is taking the msg. Translate it to its ERP
  • ERP Y is taking account of the message

My conviction is blockchain could eventually replace steps 2 to 4 , cheaper and more efficient. Creating an EDI universal station on the blockchain would be an enormous bargain !

2 comments

Lisk roadmap makes it clear that within the next step called Network Consensus, LIP23 will be implemented. Before this, need to be launched "Security and Reliability" and "Network Economics". Let's say it is not before July 2020 on current speed of devs.
The current DPOS protocol, where any lisker can vote 101 times, will be replaced by LIP23 where 1 lisk can be used to vote only on 1 delelegate. The delegate must also vote for himself with a minimum of 10% against external votes. I inviting you to read the entire protocol. It is facinating , simple but complex, justice and diabolic....

New Era will apply to enter in this game with delegate liberspirita, currently owning up to 550000 lisk, enough for the selfvote requirements and also to add external votes on some new and skilled delegates. New Era will continue to act as a pool, with revenues shares for voters, selfvoting excluded. The list of Goods/Bads may be slighly different, perhaps without Bads (I do hope it won't be necessary). But list of Goods could still be maintained to push new delegates , with activation of the 4th rule of NewEra ; voting on the weakest Good will give to the voter an additionnal bonus on rewards. Others rules may not change : voting on any of the Goods gives right to rewards, even if liberspirita himself is not voted. But rewards paid only if liberspirita is in forging position.

If several "friends" are using same protocols as NewEra, constituting a soft group, voters will also receive rewards from several delegates with only one vote. Soft groups differ against current groups because at any time it will never a mandatory to vote all, anyway LIP23 is clearly avoiding it.
Pulling a friend into 101 could be also done more actively : a lending of funds may be necessary so that the "weak friend" could assume his sefvote requirement.

No comments

Those who know me already know my response. It is NO of course. But this is not enough. Since i am interesting in cryptos, coders are always pretending to lead it. It is as silly as letting a military man leading a war or a mechanical competing on formula one. Coders have skills we need, they can code.OK for that.
I work with coders, daily. And you know what ? If you leave a coder in total freedom, he is coding only sweets or shits for no use. Tools for coders, tools for improving the tool of another coder, etc .
But when it comes to go on real world, they are lost. I need every day to improve industrial processes, financial flows, logistical issues. If i do not give to coders deep analysis, nearly with the lines of code inside, sql queries, tables involved, risks to avoid, tricks to investigate, the coders are lost.
Even in Lisk community, many so called 'leaders' are essentially coders without large scope of their possibilities. Many are collecting their 'projects', a long list of swiss tools never used (some are very good), but very rarely a project where real world would be connected to Lisk.

So i tell you : Lisk needs delegates from every where with every skills. Up to you ;)