On Post, @kathygriffin challenges Noam about Andreessen and he starts backing away (see screenshot). To which Kathy responded:
"Please don’t try to gaslight anyone on Marc Andreessen, of all people.
"Please understand that real world people had no clue about who Elon Musk and Peter Theil were, much less the destruction we didn’t catch in time. You didn’t HAVE to go to AH. You could’ve self funded."

@kathygriffin @jeffjarvis I don’t understand why anyone would want to be on another platform with an algorithm. It sucks.

@Michele @kathygriffin @jeffjarvis

I disagree. An algorithm is not intrinsically bad. As long as we understand that it represents the interests of whoever paid to have it constructed. I think an algorithm with human values that simply wanted to enrich experience is perfectly possible.

I haven't seen one, probably because nobody has ever had a financial incentive to construct it.

Mastodon would be a good place to try to make one.

@kathygriffin @Michele @timbray @jeffjarvis I could not disagree with anyone who wants an algorithm more. No algorithm is part of what makes Mastodon differ from the other social networks. Let people decide for themselves what they want to see on their timelines.

@Sanlear @kathygriffin @Michele @timbray @jeffjarvis Amen - part of hat I've grown to HATE about Facebook is that it prioritizes image and video over text when sometimes the most important updates c om far-flung friend and family networks are text. Instagram prioritizes frequent posters when the ones I want to see (agaim, f&f) don't typically post daily. IF an algorithm is created it must be fully customizable, including the ability to completely opt-out.

@jeffjarvis @kailagenevieve @Sanlear @timbray @kathygriffin @Michele people who are against algorithms can just turn them off in their settings. even twitter allows users to go back to chronological feed. There’s a reason most people don’t choose that

@kailagenevieve @timbray @kathygriffin @Jslez @jeffjarvis @Michele Maybe it’s time for people to learn better then.

This isn’t Twitter or Facebook. People here don’t need algorithms force feeding them content. We can make this experience what we choose it to be. There’s a powerful freedom in that.

Algorithms, in my opinion, are everything that’s wrong with social media.

@Sanlear @kailagenevieve @timbray @kathygriffin @Jslez @jeffjarvis @Michele

#Productmanagers and #UX folks at #BigTech clearly found that people prefer "algorithms," otherwise they would have never implemented them. PMs at companies that big make decisions based on data.

The key question for #mastodon, however, is whether to take a paternalistic stance of "we know algorithms are unhealthy for you" and force them to "eat broccoli." It's a question of what behavioral #economists call "nudging."

@rachelwilliams @Sanlear @kailagenevieve @timbray @kathygriffin @Jslez @jeffjarvis @Michele

Agree, as with "AI/ML" the value of the outcome entirely depends on what outcomes you want. There are orderings of my timeline that I would like that aren't focused on increasing ad revenue.

* Grouping and ordering thread
* Presentieg conversations between people I follow chronologically
* Showing me toots from people I value that happened since I last looked

Lots of options here

@andy @rachelwilliams @Sanlear @kailagenevieve @kathygriffin @Jslez @jeffjarvis @Michele

Wow, I slept in a bit (west-coast time) and woke up to this mega-thread. Thanks for all the smart remarks.

I got sufficiently exercised that I wrote a blog piece trying to explain why I think algorithms are necessary and what we might expect from them: tbray.org/ongoing/When/202x/20

I stole the title from Jenny Holzer.

@timbray @jeffjarvis Why is everyone focused on algorithms as *the* way to sort and filter content on Mastodon? I understand that was Twitter’s MO, but user-editable rule-based tools would probably make way more sense, especially since:

1. Personalization algos present serious privacy challenges.
2. Algorithms are much harder to work with than rules/heuristics.

@timbray @jeffjarvis Rules are executed as structured database queries (I.e. simple SQL). This approach is not “procedural” in the way that an algorithm is.

@jordan @jeffjarvis Heh, all us computer geeks are convinced that if a computer takes some data and re-arranges it to produce some particular result, that means an algorithm is being executed.

Pretty clear that the word "algorithm" has taken on a new popular meaning. No point in pushing back against that but maybe we geeks do need to invent some new terms so we can talk about filtering/curation tech without making people think we're reinventing Facebook…

@jeffjarvis @timbray Agree and I don’t want to get buried in CS semantics, but the most important point is that we can have great content filtering without complicated mathematical processes. This filtering can be executed as simple search queries that anyone can edit.

@jeffjarvis @timbray And, as a side note, the approach I’m outlining is far less computationally expensive than typical content recommendation algorithms. This matters in a FOSS environment much more than in a for profit enterprise.

@jordan @jeffjarvis Turns out that high-volume content filtering has been the focus of most of my actual hands-on technical work in recent years and you're right, rule-based systems are really a good way to go, can be made insanely efficient.

@timbray @jeffjarvis Want to hack on a prototype where you can set up simple standing Mastodon DB queries?

@jordan @jeffjarvis maybe. Current energy going into getting cosocial.ca launched.

@timbray @jeffjarvis Very cool. One thing that Twitter never got right was that their power users are deeply passionate about specific topics. The tools they provided (lists, search, and hashtags) were always deeply flawed and prone to junk. There’s a better way!

@jordan @jeffjarvis @timbray A couple of thoughts on this: I'd check your definition of "anyone." There are a lot of people for whom even the simplest "advanced search" language is just way too much for them to grok. There *is* potential user-experience value in having an AI/ML algorithm that can learn from what you like/dislike and give you more of the former and less of the latter over time.

@jordan @jeffjarvis @timbray This is not to defend FB/$8chan's algorithmic optimization for eyeball monetization. And you're absolutely right that rule-based filters are probably more effective/computationally efficient for smaller instances. And any algorithm/filter/etc. should certainly be optional for the user and transparent to the extent possible. All I'm saying is that I wouldn't *entirely* rule out AI/ML algorithms as a valuable way to manage content.

@timbray @aelman @jeffjarvis I hear you. I’m not anti-recommendation algorithm at all (I’ve worked on many). My pitch is that we build simple UI tools to build filters, so truly anyone can use them. Reddit has some of these and they seem to be widely adopted.

@jordan @timbray @jeffjarvis Makes sense to me! And of course, one of the beautiful pieces of a federated model is that if someone wants to come in and create a service - OSS or proprietary - that uses an AI/ML recommendation algorithm to filter/order content, they can do that, and end users can choose what they want without sacrificing access to the people/content they want to follow.

@timbray @jordan Yes, please. The word "algorithm" (not to mention the word "data") has been demonized.

@jeffjarvis @timbray @jordan

I suggest the word "function" because we're going to start a new Wikimedia project, Wikifunctions. So I want that to be a good word.

Sign in to participate in the conversation

Everyone is welcome as long as you follow our code of conduct! Thank you. Mastodon.cloud is maintained by Sujitech, LLC.