It's not so much that people forget about Usenet as that they forget why it died.

It got spammed to death.
It lost control over its culture, and that culture was crucial to its functioning.
It was too problematic for ISPs (or others) to provide ready access to it: spam, harassment, child pornography, and copyright violations all posed massive concerns.
There was no viable business model for providing the service.

https://redd.it/3c3xyu

#dreddit

@dredmorbius I've tried to explain Eternal September to people who came online later, and they can't quite comprehend. But it's a problem every service undergoes if it's successful.
@dredmorbius When you mention Usenet's "suggested successors" are you thinking Reddit and the chans which to some extent follow its organizational principles, or are you thinking of projects more like Mastodon, which to some extent follow its architectural principles?

@Brook Ultimately, more the social-media elements than the technical-media elements, but the medium is the message, so they feed off each other tremendously.

My memories of Usenet start in the mid-late 1980s, when it was a small, mostly academically-selective community of a few thousand people, also coloured by my own youth, immaturity, and ignorance. But it left a warm sense of a place I've wanted to got back to. Of course you never can.

I've thought a lot about what made it good and bad.

@Brook Ultimately, the content and dynamics are more important than the mechanisms, but since one feeds the other, both matter.

There's also who's included, who's not, on what basis, and a mess of other stuff.

I make heavy use of Reddit, none of the chans, various other networks over the years: #slashdot, #kuro5hin, #hn, #googleplus, #ello, #imzy.

Some have worked well, some not. (Imzy is an excellent case of the latter.)

@Brook One thing Mastodon / OSocial are offering that I've looked at for a while is the idea of a /locally-organised/ community, where "local" isn't (necessarily) based on /geographical/ space, but there's the idea of places nearer and further from others within the network.

On Mastodon presently, you could draw a hierarchy of distances, some official, some more logical / semantic:

Local -> MastodoNet -> OSocial Fediverse -> Social networks (Birdland, FB, G+) -> Web -> Internet -> Universe.

@Brook Keep in mind: that's not a straight-line expansion per se, but it's hard to draw multidimensional graphs in 500 chars.

@dredmorbius I was late to Usenet, probably getting started on it no more than a couple months before the fateful September. Casual Reddit user, and occasional chan lurker out of sociological interest. Fertile ground for emerging culture, because it's heavily fertilized.

I keep thinking my perfect social service would merge timeline from Twitter, auto-expiring posts from chans, single sign-on + anon accounts from Reddit, and a loosely-coupled long-form capability from Medium.

@Brook Good points all.

Keep in mind that all data expires. It's just poorly specified in most systems.

@dredmorbius That's definitely true. I'm thinking a defined public timeline expiration period, say 36 or 48 hours, combined with the option to easily save to a clipping folder only viewable by participants in the clipped thread. Objects (whether posts or threads) in the clipping folder could be reactivated by reposted to the public timeline, as long as all participant accounts are active and haven't blocked the clipped object.

Clearly, I've thought more about this than even I realize...

@Brook Nice. I'm liking. Might argue the timeframe, but it's a good start.
@dredmorbius Would you argue longer or shorter? It seems to me one of the problems with shorter is it would encourage people to repost *a lot* for visibility. Better to let things linger and put some thought into discoverability instead.

@Brook Longer. At least a week, maybe 3-6 months.

If it's not been found by then, expire.

CDNs may have good guidence on cache hit rates.