As Twitter is ruminating about deleting inactive accounts again, I want to share here about a proposal which was part of a university thesis, with some improvements based on current context.

The inactive accounts, in this proposal, will be moved to a cold archive after a prolonged period of inactivity, ranging from 6 months to two years. The cold archive consist of:

1. A separate subdomain from main Twitter. Example archive.twitter.com or limbo.twitter.com

2. While accounts retain their original usernames as part of archive usernames, their mainspace usernames will be freed up. Discord-style number discriminators may be used so to prevent archive username exhaustion, although keeping their names intact during the process is ideal.

3. If an archived account is accessed, there will be a prompt to ask whether to restore to mainspace or leave it as it is. In the case of the former, their mainspace usernames are usually restored, but if someone else has taken up the slot then they have to choose a new username.

4. Past search histories that are related to the account is deleted upon archiving.

5. While as in normal mode, tweets and users can be browsed through its internal search engine, a robots.txt will forbid Google and other bots to index any and all contents of the archive subdomain. Exceptions only apply to Internet Archive's bots.

Edit: In response to the frame challenge below, this is my take as to why. Originally posted as a reply comment but somehow it didn't show up.

Many old Twitter accounts have value, either objective or subjective/sentimental (deceased relative/friend). Besides, there are a lot of contents, tapestries and so on that accumulated at Twitter over a long time recording the evolution of culture in general.

If they move forward with the purge the following will be lost:

1. Accounts whose owners are alive but not be able to log in for any reasons, such as being on military active duty, jailed (can be wrongful - look up the Innocence Project), or under total or partial internet blackout like Myanmar or Kashmir.

2. Tweets from now deceased users related to major upheavals such as the Arab Spring - extremely important and valuable for anyone looking into the topic.

3. Fictional works. Often there are so many "roleplay" accounts or otherwise which was active for a short duration just to act out of its fiction, but has since gone dormant.

I can't but to think of Margaret Atwood's "The Testaments" especially about the "loss of internet history in the latter chapters set in a 22th century symposium.