[lilyball]

They haven't scrapped their plans for this. They said they'll put it on pause until they've come up with a tool for memorializing accounts. So presumably accounts that are legitimately not going used will still eventually free up.

Assuming of course that the account is truly idle and not being used to read without every interacting.

[duskwuff]

A simple fix would be to make an exception for accounts with >X statuses posted or >Y followers.

The inactive accounts that people are worried about losing are the ones that used to be active, or which were significant enough to make a lot of people follow them. However, the bulk of inactive accounts probably have very few (<10) statuses posted or followers, and deleting those is likely to be uncontroversial.

[tom_mellior]

A simple fix would be to pick a character that can not otherwise occur in Twitter handles, like ':'. Then move every inactive @foobar user to @inactive:2019:foobar. If a follower of the original account complains, change "inactive" to "memorialized" or whatever, possibly adjusting the year backwards.

If I choose to squat the now freed up @foobar handle and not do anything with it, next year they can move my inactive version to @inactive:2020:foobar and free up @foobar again for the next person interested in using it.

Any way they solve this, there is a danger of people using bots to mass-squat liberated names to offer them for sale. I don't see any very good way around this. Possibly to give first pick to users with long-standing accounts that share a nontrivial common substring with the liberated handle. For example, if my current Twitter name is @firstname_lastname, and I really want @firstname, I think it might make sense to prioritize me over @random_bot_204safhq23.

[lilyball]

Twitter's in a good position to detect most bots. If someone uses a bot that's good enough to evade detection, I think it's ok to let them keep the handle.

[teddyuk]

My relatives who died have fewer followers than your relatives who died so yours stay and mine go? Doesn't sound fair!

(my relatives also didnt tweet very often but when they did they were amazing)

[kkarakk]

I can understand not making a tool to memorialize users in first 3 years of creation but once twitter became a household name they really SHOULD have already updated their roadmap.

In the era of cloud computing, how expensive would it be to make an account readonly and store the paltry data it created in a cached location somewhere?

All of their tech decisions feel really reactive nowadays

[L_Rahman]

With roadmap decisions like these, the compute or storage cost for the data is an insignificant part of the product decision making.

The real cost of doing something like this is doing user research to figure out the set of human behaviors around the memorialization process, deciding where in the product to make the functionality available, training material for how it escalates up to human support teams, and what other things are on the team's plate at the same time.

[austhrow743]

Why does it have to be specifically the tech holding them back from doing it?

Personally the permanence of internet content is terrifying and by far the worst aspect of it. Potentially there are people like me who would be at the table whenever "what to do with accounts of dead people" is discussed, strongly advocating for deletion. Lacking clear consensus they just left the decision until it had to be made.