5
  1. If you make a new wallet, (seed), and move all your tokens to it, will the other (old) wallet disappear?
  2. Will the old wallet (seed) clog the tangle?

  3. Is there a preferred way?

  4. Is there ever a time/situation where we will HAVE to generate a new seed. (Except because of improper address reuse prior to a snapshot -- for recovery purposes).

  5. In other words can our seeds remain permanent if we follow the present best practices?
Helmar
  • 1,293
  • 1
  • 15
  • 28
til
  • 95
  • 4

1 Answers1

5

No you don't ever need a new seed when you follow the present best practices.

The only reason to need a new seed is if you lose your old one or if your seed gets compromised—as in someone else knows it.

The seed does not get compromised through address reuse. Only the private keys can get compromised. They are derived from the seed. More in this answer. Regarding the first few points, the seed is never in the tangle. Thus, it can neither disappear or clog the tangle.

Helmar
  • 1,293
  • 1
  • 15
  • 28
  • At the moment, it might be worth to change seed after a significant number of transactions through. The current light wallet downloads all transactions after entering your seed, if you have a lot (>50) of them, this might take a little longer to load – Werner der Champ Jan 04 '18 at 20:09
  • Thanks, that's the answer I was looking for. As an aside you might want to substitute Winternitz for Lamport in your link. – til Jan 04 '18 at 20:16
  • @WernerderChamp, good to know. Although that's just suboptimal software design. – Helmar Jan 04 '18 at 20:17
  • @til that's not my answer. Why should they be replaced? – Helmar Jan 04 '18 at 20:19
  • Werner, do you think zero balance transactions will be eliminated (without a snapshot) in future wallet versions? Is it possible/probable that consolidation of sites may be forthcoming where multiple transactions are consolidated into one site? Thanks – til Jan 04 '18 at 20:20
  • Sorry Helmar, I thought I was commenting on YOUR answer. Also sorry I missed 'variation of' in your link "Iota use a (variation of) the Lamport algorithm." Thanks again – til Jan 04 '18 at 20:23
  • @til you did comment on my answer. The linked answer isn't mine though ;) – Helmar Jan 04 '18 at 20:27
  • Helmar, if you want to feel the lag enter "A9999999...999" as a seed ;) @til That would be pure guesswork. I could imagine that nodes will wipe the message from zero-transactions for more efficient storage management. The transaction itself (branch tx, trunk tx, address) would stay through – Werner der Champ Jan 04 '18 at 20:30