Papers
categories: reading
tags: paper
Description/Summary
Papers I’ve read.Content
VM Matters: A Comparison of WASM VMs and EVMs in the Performance of Blockchain Smart Contracts
Determinants of Early-Stage Startup Performance: Survey Results
To explore determinants of new venture performance, the CEOs of 470 early-stage startups were surveyed regarding a broad range of factors related to their venture’s customer value proposition, product management, marketing, technology and operations, financial management, funding choices, team management, and founder attributes. Multivariate regression analysis shows that startups that employed lean startup management practices—in particular, an optimal rate of pivoting—had stronger seed equity valuation growth, as did new ventures that struck the right balance in recruiting for skill versus attitude and that professionalized human resource policies earlier. High levels of confidence in financial metrics—including estimates of unit economics, the ratio of customer lifetime value to customer acquisition cost, and total addressable market size—were also associated with strong growth in seed equity value.
Valuation outcomes were not related to several founder/CEO attributes, including age, educational background, personality traits, and motivations for becoming an entrepreneur. Likewise, choosing angels versus venture capital firms as lead seed round investors did not predict subsequent growth in equity valuation, nor did decisions about partnerships to provide technology, operational capacity, or marketing support.
Aggregation of Gamma-Signatures and Applications to Bitcoin
Not a great paper. I don’t think it has strong proof of its point, and I am not sure if it’s workable as their reasoning.
Aggregate signature (AS) allows non-interactively condensing multiple individual signatures into a compact one. Besides the faster verification, it is useful to reduce storage and bandwidth, and is especially attractive for blockchain and cryptocurrency.
Currently, due to the 1M-byte limitation of block size, about 7 transactions are conducted per second in the Bitcoin system. This leads to, in particular, longer confirmation latency, relatively higher transaction fees, and easier target of spam attacks.
Bitcoin uses the EC-DSA signature scheme [37] over the secp256k1 curve [22]. According to Bitcoin Stack Exchange, in a standard “pay to public key hash” (P2PKH) transaction or a “pay to script hash” (P2SH) transaction, the signatures occupy about 40% of transcript size. In addition, an EC-DSA signature involves non-linear combination of ephemeral secret-key and static secret-key, which is the source for relative inefficiency and for the cumbersome in extensions to multi-signatures [8, 47], scriptless scripts [66], etc. As a consequence, recently there is also renewed interests in deploying Schnorr’s signature with Bitcoin in the future.
Practical aggregate signature schemes were proposed [17, 7] in the plain public-key model. They are derived based on the BLS short signature [18] in groups with bilinear maps. There have been some discussions on deploying the pairing-based AS schemes [17, 7] in the Bitcoin system [46], which are briefly summarized below. – System complexity. Deploying pairing-based aggregate signature schemes requires the replacement of not only the EC-DSA algorithm but also the underlying elliptic curve. It makes a deployment in practice (such as Bitcoin) much more invasive than simply shifting algorithms. – Bilinear group vs. general group. Intractability problems in groups with bilinear maps are weaker than the discrete logarithm problem in general EC groups. – Verification speed. As an individual signature scheme, the verification of the pairing-based BLS signature is significantly slower than that of EC-DSA. Note that the miners still need to verify the correctness of individual BLS signatures before aggregating them into a block. Some survey indicates that on a concrete hardware it can verify 70,000 secp256k1 signatures per second, while it could only verify about 8,000 BLS signatures per second [46].
It is thus highly desirable to develop aggregate signatures, with the following features simultaneously: – It can be built from general elliptic curves (without bilinear maps), in the plain public-key model with fully asynchronous communications. – The underlying signature scheme has provable security, and moreover, is more efficient and flexible than EC-DSA.