fix research.web3.foundation links

docs/id-weights.md

@@ -750,4 +750,4 @@ next\ weight &=& weight \times (1 + (v \times diff) + (v \times diff)^2 / 2)
 \end{aligned}
 $$
 
-Polkadot defines the `target_weight` as 0.25 (25%). More information about this algorithm is described in the [Web3 Foundation research paper](https://research.web3.foundation/en/latest/polkadot/overview/2-token-economics#relay-chain-transaction-fees-and-per-block-transaction-limits).
+Polkadot defines the `target_weight` as 0.25 (25%). More information about this algorithm is described in the [Web3 Foundation research paper](https://research.web3.foundation/Polkadot/overview/token-economics#relay-chain-transaction-fees-and-per-block-transaction-limits).

docs/sect-block-production.md

@@ -212,7 +212,7 @@ It is imperative for the security of the network that each block producer correc
 
 Polkadot does this synchronization without relying on any external clock source (e.g., through the or the ). To stay in synchronization, each producer is therefore required to periodically estimate its local clock offset in relation to the rest of the network.
 
-This estimation depends on the two fixed parameters ${k}$ ([Definition -def-num-ref-](sect-block-production#defn-prunned-best)) and ${s}_{{{c}{q}}}$ ([Definition -def-num-ref-](sect-block-production#defn-chain-quality)). These are chosen based on the results of a [formal security analysis](https://research.web3.foundation/en/latest/polkadot/block-production/Babe#-5.-security-analysis), currently assuming a ${1}{s}$ clock drift per day and targeting a probability lower than ${0.5}\%$ for an adversary to break BABE in 3 years with resistance against a network delay up to $\frac{{1}}{{3}}$ of the slot time and a Babe constant ([Definition -def-num-ref-](sect-block-production#defn-babe-constant)) of ${c}={0.38}$.
+This estimation depends on the two fixed parameters ${k}$ ([Definition -def-num-ref-](sect-block-production#defn-prunned-best)) and ${s}_{{{c}{q}}}$ ([Definition -def-num-ref-](sect-block-production#defn-chain-quality)). These are chosen based on the results of a [formal security analysis](https://research.web3.foundation/Polkadot/protocols/block-production/Babe#5-security-analysis), currently assuming a ${1}{s}$ clock drift per day and targeting a probability lower than ${0.5}\%$ for an adversary to break BABE in 3 years with resistance against a network delay up to $\frac{{1}}{{3}}$ of the slot time and a Babe constant ([Definition -def-num-ref-](sect-block-production#defn-babe-constant)) of ${c}={0.38}$.
 
 All validators are then required to run [Median-Algorithm](sect-block-production#algo-median-algorithm) at the beginning of each sync period ([Definition -def-num-ref-](sect-block-production#defn-sync-period)) to update their synchronization using all block arrival times of the previous period. The algorithm should only be run once all the blocks in this period have been finalized, even if only probabilistically ([Definition -def-num-ref-](sect-block-production#defn-prunned-best)). The target slot to which to synchronize should be the first slot in the new sync period.