Denis Reshetov / E-mail: [email protected] / Telegram: @denisreshetov
Reputation is the cornerstone of economic life and is essential for the fast growth. Here I describe a novel approach to reputation digitalization that is discrimination resistant, simplifies hiring process and revolutionize education in schools, universities and at companies.
Recommendation letters is a well-known decentralized approach to manage reputation that has proven to be effective and stable for more than centuries.
A referee issues recommendation about a worker that can be shown to an employer during an interview. Unfortunately such traditional recommendation letter lacks responsibility of the referee. An evil referee can issue a fake recommendation that will damage the employer. This fact limits usage of recommendations during hiring and leads to an excess of time that workers and HR specialists spend on job interviews: the same skills are reexamined multiple times in different companies just because employers can't fully trust referees.
I propose to split a person's reputation on two parts to solve this problem and make recommendations fake resistant.
The reputation of a person should consist of two parts:
- Electronic recommendations letters issued about the person, backed by referees SLON balance
- SLON token balance of the person as a referee
With such approach an employer can slash SLON tokens of an evil referee if a worker lacks skills mentioned in a recommendation letter. The employer can sell such SLON tokens on an open market to compensate HR specialist time lost for the reexamining of a fake recommendation.
There is a probability that the good referee will be pennalized in a case if the worker will forget skills mentioned at the recommendation. As in insurance business model the worker should pay a premium to the referee, otherwise the referee will always lose money in long term.
The described scheme was initially developed at Slonigiraf private school to motivate teachers work well and eventually leaded to a p2p-learing model: every student can be a mentor and issue a recommendation about his mentee, the mentee pays the mentor for teaching and issuing a recommendation. The senior teacher buys recommendation letters about mentees by providing them A-grades (or money). It seems that such approach can be a great option to rebuild all types of education.
The proposed reputation system is resistant to a Big brother surveillance and any discrimination:
- All recommendation letters are stored privately on personal devices and showed only to organizations that the person trust
- SLON token balance can't be transformed in a discriminative Social Credit System (ex. Chineese one) because anyone can buy SLON. If a criminal buys SLON and shows high SLON balance it doesn't make him good or socially acceptable. Moreover accounts are anonymous and a person can have multiple accounts to store SLON tokens.
Insurances are needed to protect from financial loss. Recommendations backed by SLON are insurances in fact.
An organization which provides insurance known as an underwriter. A person who buys insurance is known as a policyholder, and a person covered under the policy is called an insured. In case of SLON recommendation letters the refeeree is the underwriter, the worker is the policyholder and the employer is the insured.
The insurance creation requires a payment from the policyholder to the insurer known as premium in exchange for the insurer's obligartion to provide a reimbursement to the insured in the event of a covered loss. SLON recommendation letter creation also requires the premium which is paid by the worker to the referee.
A contract provided to the policyholder is called the insurance policy, which explains in which case the insurer will compensate the insured. The recommendation letter is the the insurance policy.
Note, that the worker can optionally charge money from the employer for the insurance policy (recommendation letter) usage.
In a case of a loss the insured submits a claim to the insurer for processing by a claims adjuster. The claims adjuster determines the extent of the underwriter liability. In case of SLON recommendation letters the employer itself is a claims adjuster. However the employer can't do unfair decisions because it will ruin its public creditability and hiring process: no new workers will wish to provide recommendation letters to such employer.
| Insurances | SLON recommendation letters |
|---|---|
| Underwriter | Refeeree |
| Policyholder | Worker |
| Insured | Employer |
| Insurance policy | Recommendation letter |
| Premium | A payment from a worker to the referee for a recommendation creation |
| Covered loss | Employer's financial loss due to a lack of skills of the worker |
| Claims Adjuster | The employer itself monitored by the worker and the refeeree |
| Reimbursement | SLON token transfer from the referee to the employer |
Table 1. SLON recommendation letters in terms of insurances
Figure 1. SLON and recommendation letter markets connection
- A refeeree buys SLON tokens to issue recommendations
- The refeeree sell a recommendation letter to a trusted worker about the worker
- The worker sell (or transfer for free) the right to slash the referee to an employer during interview to get a job
- In case of the fake recommendation the employer slashes the referee and sell SLON to compensate time of HR specialist lost for interviewing the worker that lacks skills mentioned in the recommendation letter
Paid interview is a known instrument to recruit a specialist when there is a deficit of qualified candidates. The recommendation letter market can save a budget for paid interviews because employers can get a reimbursement for a weak candidate.
TODO: write this section about mentors and mentees
Education is lagging behind the changes in the world. Companies need new skills, and educational organizations don't have time to teach them. Often, education does not even know about their relevance.
We are solving this problem with the help of the market. The market always knows what skills are needed, how to teach them and for whom they are needed. In modern markets, the value is not the product itself, but the guarantee of its quality, thanks to which you can replace the product or get money.
According to the Slonigiraf system, a student buys a skill guarantee from a teacher and resells it to the company they work for or to a philanthropist who finances the student's education. The guarantee allows them to take an asset from the teacher, SLON coins, in the amount that is recorded in the guarantee, if the student's skill disappeared. SLON coins play the role of a teacher's digital reputation that can be bought and sold.
Reputation is a measure of public confidence. Using reputation allows you to reduce costs in the economy, trust is profitable. However, reputation cannot be used if a person changes their social circle. In addition, a reputation is slow to respond to mistakes and oversights of its owner.
Digital reputation is devoid of these problems, but it has a different negative side. For example, in China, where the government is the operator of the reputation base, you may be denied access to public transport, issuing loans, etc. if you have a low digital reputation. At the same time, you can increase it only by changing your behavior in a way pleasing to the state.
The digital reputation of SLON is free from this drawback and does not lead to discrimination of the owners, since SLON is available for purchase.
Anyone can use a SLON reputation to issue a guarantee for another person's skill. Because the issuance of a guarantee can lead to a decrease in reputation, the guarantor asks for a certain amount of money from the recipient of the guarantee in exchange for its issuance.
Most often, teachers act as guarantors, and students buy guarantees in the hope of reselling them to future employers or patrons.
To issue a guarantee, the teacher must create an account within the Slonigiraf system and transfer the required amount of SLON there. The number of SLONs on this account is known to all Slonigiraf users. Thanks to this information, they know that the guarantee can be used and the amount of SLON specified in the guarantee can be debited from the teacher's account. SLONs on the teacher's account are not blocked in any way. You can, for example, put 100 SLON on your account and issue 5 guarantees, each of which will allow you to fine the teacher by 100 SLON. After the execution of the penalty on the first guarantee, the account will be reset to zero, and everyone will know that the other 4 guarantees are not valid yet. If the teacher refills his account with 100 SLON, these 4 guarantees will become invalid again. An example of the application of the Slonigiraf methodology in a public school can be seen in Fig. 2.
Figure 2. An example of the application of the Slonigiraf methodology in a public school
At the moment, we know 4 examples of using this educational format in practice: Family school Slonigiraf (Moscow), Group study of biology online without a teacher, In the work of tutors, Teaching the values of SLON.
Family school Slonigiraf (Moscow)
The entire educational format of Slonigiraf was conceived in a private family school while looking for a way to motivate teachers to work for the result. At Slonigiraf, children buy skill guarantees from teachers and resell them to their parents. Children themselves can act as teachers. The quality of this interaction is monitored through guarantees. Poor skills are exposed by parents who penalize teachers, as well as by teachers themselves, who can double-check student guarantees from previous teachers. Teachers do not receive any wages, they earn only by selling guarantees.
In public schools and universities
We have created a paper version of the Slonigiraf methodology so that teachers can quickly try out this methodology in their lessons. The teacher just needs to follow the link: https://slon-i-giraf.ru/app/work?view=paperGameView&language=ENG
write 3 skills that you need to pass in the lesson, click on the "download" button and print the resulting file. This file contains step-by-step instructions on how to conduct the lesson.
So far, this version of the methodology has been used by teachers in 6 public schools in Russia.
Online group study of biology without a teacher
Every week, more than 30 high school students gather for an online meeting at Zoom in order to issue guarantees to each other on biology skills. At each of the meetings, they are divided into teacher-student pairs, the teacher asks the student for half an hour on a selected topic from biology and issues guarantees. Then, inside the pair, they change roles. The quality of teaching is controlled by the fact that every week the student selects a new teacher, who selectively checks the guarantees issued by the previous teachers. Coordination of meetings takes place in the Telegram group
In the work of tutors
Tutors issue guarantees for the skills they have taught their students. Pupils sell these guarantees to their parents for money. The parent can fine the teacher if the skill is lost. When the tutor runs out of SLONs, he buys them from the parent.
SLON value training
To make people learn how to use Slonigiraf in teaching, we buy guarantees for the "SLON Value" skill. Teachers who understand our format, teach their friends, issue them a guarantee, in exchange for the student's promise to pay off SLON. New users receive 600,000 SLON for this skill from the creators of Slonigiraf. Students give 200 thousand SLON to their teachers for writing a guarantee. You can get SLON in the Telegram group of the project
In order for the Slonigiraf system to be used to train a large number of people, we need to process a huge number of transactions per second.
We chose a well-known network architecture Polkadot, which consists of computers storing a base with SLON balances - a relay chain, and computers storing a guarantee base - parachains. The main load falls on parachains, scaling occurs due to the fact that hundreds of guarantee bases can be created, for example, one for each country.
The relay chain also protects the parachains from attacks.
- SLON balances, transfer and receipt are available on the page:
https://polkadot.js.org/apps/?rpc=wss%3A%2F%2Fwss1.slonigiraf.org#/accounts
- Issuance of guarantees and their verification:
https://slon-i-giraf.ru/app/work?language=ENG
- Project news group:
- Chat:
- A student finds a teacher - somewhere offline based on a knowledge field, personal considerations and amount of SLON at the teacher's escrow account.
- The student and teacher trade with each other on the price of insurance issuing during this session and a number of insurances that the student gives to the teacher for revalidation.
- The student sends a list of insurances to revalidate - they should be linked to a teacher account with escrow that balance is more than possible fine.
- The teacher revalidates them and fines previous teachers for invalid insurances.
- The student sends a list of the skills to be discussed at the session to the teacher
- The teacher marks at his own device which skills are good and which are not.
- The teacher tells the student an amount of SLON to pay for total insurance on good skills.
- The student transfer the required amount of SLON to teacher personal account
- The teacher signes with his private key a hash (1) that contains a long random number concatenated with the skill ipfs hash with an address of the student and separately the same just without ipfs hash.
Each such paired data - is one insurance. And transfer these to the student.
Secret skill:
skill_ipfs_hash | insurance_id | teach_address | stud_address | amount | teach_sign_1
Skill receipt:
insurance_id | teach_address | stud_address | amount | teach_sign_2
Skill info:
skill_ipfs_hash | insurance_id | teach_address | stud_address | amount | teach_sign_1 | teach_sign_2
Skill insurance:
insurance_id | teach_address | stud_address | amount | teach_sign_2 | employer_address | stud_sign
insurance_id - u32 value
Ex: 5
teach_address, stud_address, employer_address - substrate accounts.
Ex: 5GrwvaEF5zXb26Fz9rcQpDWS57CtERHpNehXCPcNoHGKutQY
Both skill receipt and skill insurances can be sent to the blockchain.
transaction_id is stored at map containing teacher info
- The student goes to the company. The company buys the: part2 of insurance concatenated to the company address + sign of it by student private key. Thus the company can send an instruction to blockchain: send money from teacher to company. The information stored at blockchain: address of insurer, address who gets a reward, the amount and hash (1).
used insurances are stored in map
H256 -> H512
where H256 = hash(Identity+insurance_number/512)
H512 = 1...0…
1 - for used insurance, 0 - free