112

Regarding the reversal of the AI generated content ban, once again I'm astonished at the self-serving decisions and actions taken by those who profit materially from this network. Alienating trusted and invested users, be they moderators, high-rep users with expertise in the field(s) they post in, users donating their time by voting, answering, flagging (etc.), encouraging new users, whatever, by clearly putting profit before quality cannot possibly help the network to thrive, maybe even to survive. It will certainly result in a loss of respectability as a source of high-quality and trustworthy answers.

What is so difficult to understand about the fact that you (the SE network) need us to survive, but that we users don't need you to do so?

The fact that we're seeing another strike due to actions harmful towards those who contribute voluntarily so soon after the last fiasco leaves me (almost but not quite) dumbstruck. It took an unhelpful likely AI generated answer (now deleted) to compel me to post here. (This is the user's first post. Unusual for many, this user has actually added info in their profile wherein they state,

Through my passion, expertise, and relentless pursuit of excellence, I am poised to emerge as a future leader and innovator in the field.

How much more ironic can this situation be? AI masking as expertise.

The people who make this network valuable are not the people who post AI generated answers, it is those with actual knowledge (including but not limited to expertise), the altruistic desire to help others, and the willingness to donate some of their time to answer, moderate, edit/improve answers, down vote and delete bad answers, etc.

I'm a physician. That is considered by some (especially physicians) to equate with expertise in the field of medicine. I tried to moderate on what is now called Medical Sciences. I used to moderate Parenting. The first time I quit was over the last fiasco. I ran for moderator relatively recently out of concern over the fact that the other candidates' scores were 4/40 or below, and a concern for the mods there and the site. It was not a good reason to resume moderating, but as someone who chose a profession geared specifically towards personally helping others, I kind of operate on that principle outside of work as well. I resigned again because it bothered me to contribute so heavily to this network, which I felt did not have the best interests of its valuable users at heart.

You, S.E. Inc., have placed profit before quality once again. How short sighted can you be?

Yet here I am, still. Less involved but still answering. What will it take for me to finally leave the network? Probably overcoming inertia caused in part by my inability to interact with people irl due to a severely messed up broken leg and in part by the pandemic, and that I value what I do and what I've invested here. If I can overcome this inertia, I'm outta here for good, S.E. Inc.

23
  • 10
    They have only one concern: money. Things they have no concern about at all whatsoever are: humanity, human dignity, human culture, human heritage, and importantly, consequences of their actions on humanity. That's how. Ah, and one more thing. When they will have the money, and we won't, then they will have reached their ultimate motivation: power. Look at the definition and characteristics of unbridled, pure capitalism. That's what is happening here; and it's especially offending for us, because we got personally caught up in it.
    – Levente
    Jun 10, 2023 at 15:21
  • 34
    I'm astonished at the self-serving decisions and actions taken by those who profit materially from this network -- I am astonished that you are astonished over this. The folks who made this company from nothing, most of them are no longer in the company for one or other reason. The top decision makers are "professional" people. Bring short-term results, make loads of money (well some of them do at least) and then move on to repeat. This happens in banking sector all the time (and the depositors pay the price when the bank goes down). So I am not astonished.
    – 286110
    Jun 10, 2023 at 15:23
  • 2
    It's simple, really; they think they'll be able to chug along even if they chase off a relatively small percentage of their users, even if the users who leave were making an outsize contribution to the site. The proportion of mods striking on SO is about 1 in 8, so I think in their minds, the worst case scenario is they'll have to find a few other mods to replace them.
    – kaya3
    Jun 10, 2023 at 15:24
  • 1
    Also see my comments here and here.
    – Levente
    Jun 10, 2023 at 15:24
  • 2
    Thanks to AI advancement, day by day, they hope to need us less and less.
    – Levente
    Jun 10, 2023 at 15:27
  • 46
    Much of the current leadership has never participated on any SE site. I think many genuinely don't understand how things work from the community's perspective and the implied "social contract" that has existed since its founding. There is a huge mismatch in experiences and expectations. Jun 10, 2023 at 17:47
  • 28
    I don't expect a CFO or every engineer to have an active account, but people making operational decisions that are fundamental to how the site works should probably have some first-hand experience – without it you'll never fully understand how and why the site "works" in the first place and you end up with bad decisions. No one wins with that, least of all the company. Jun 10, 2023 at 17:47
  • 1
    No mention of dupes? They don't need us. Site health actually depends on out with the old, so that it can be business as usual. Marking things as dupes discourages new users, doesn't perpetuate site traffic, nor make money, or help me get free internet points, and feels like throwing stones into a void which is the lowest form of 'interacting with people'. Which is what would make this site worth something (preserving the signal to noise as it is now, and not slipping backwards). That is however not valuable.
    – Mazura
    Jun 10, 2023 at 19:19
  • 3
    @MartinTournoij there is something more I observe: if those decision-making executives don't participate at all, and never gather first-hand experience of being here, then they will inevitably rely on some kind of second-hand accounts. In other words, the staff-members who regularly interact here, will relay a sentiment, that the executives will probably adopt. And here comes the thing: I see numerous such staff memebers interacting here who I feel are incredibly unaligned with us, and totally insensitive of and uninterested in our concerns.
    – Levente
    Jun 10, 2023 at 19:48
  • 10
    @Mazura - No. I loved ELU and am grateful that I found and participated in it. However, there were atmospheric changes that made it gradually less pleasant to spend time there, so I moved along. A more interesting story is why I very, very suddenly stopped posting on IPS, and, interestingly, it reflects poorly in S.E. Inc. as well. But I'm quite hesitant to share that story. Jun 10, 2023 at 20:27
  • 3
    "We don't need you" would be a lot more convincing with a free (perhaps even federated) alternative to the Stack software (or something superior). Even then, unfortunately, the knowledge base would have to be reconstructed. Sites like Codidact are, from what I've seen, simply not on that level. Yet. Jun 12, 2023 at 12:10
  • 4
    @KarlKnechtel - You're right, but I meant "survive" literally. I would not fall into a state of near uselessness if I left the network. If a mass exodus of users here occurred, the network would become useless relatively quickly. Jun 12, 2023 at 13:03
  • 4
    @anongoodnurse prior programmers community that I participated in was broken by management after site changed the owners (and I left it for Stack Overflow). Back then, these folks who broke it, they probably thought that they make it more efficient and improve whatever KPIs they had in their little managerial spreadsheets. Time has proven that they were just clueless and instead of the thriving site that could bring them traffic and money they now have got a bunch of dead bytes occupying their servers
    – gnat
    Jun 12, 2023 at 13:41
  • 5
    I have had a similarly difficult time leaving for good. The best that I can do is wander off for a couple months at a time now. It just makes me sad so much potential for good is being squandered and I can't even begin to see a way to fix it.
    – ColleenV
    Jun 12, 2023 at 18:31
  • 2
    @mxmissile - A product is profitable if it's useful. In this case, quality answers is the product. If profit alone is more important than the quality of the product or treating the workforce ethically, then that is short-term thinking, like shooting yourself in the gut to get out of the military - not the foot, but a gut shot - which left untreated has a pretty good chance of killing you sooner or later, unlike what you see on screen. If I didn't value the quality of my product above profit, I'd be richer, but my patients would be far worse off, and my name would be Mudd. Jun 13, 2023 at 9:56

6 Answers 6

55

I saw this quote from Alexis Ohanian

Everyone always asks me about community building and lessons learned from creating Reddit and the hard truth is that 99% of it is just showing up and doing the work of being a community leader, which means (mostly) listening, starting productive conversations, role-modeling the behavior you wanna see in your spaces. Online community-building is more like IRL community-building than people realize. Thing is — most people don't wanna do the work. And it's a lot of work.

We've not really seen organic engagement from higher levels of the company in ages, and for all intents and purposes, they're insulated from day to day community work.

Basically people don't want to do the work.

Community is a 'nice' selling point, but much like a puppy, it needs a lot more work than people realise. You need to balance the needs of various constituents - of different levels of engagement and different needs. You need to deal with people doing the 'right' things for the 'wrong' reasons. Sometimes you even need to re-examine your pet projects and ask yourself if its right for the community.

Community also has no 'book value' - it needs investment in time, energy and money that you're never going to see a direct/obvious value from unless you're talking to the people on the ground.

To make things more complicated I suspect those in power probably are looking at numbers and statistics, rather than the situation on the ground first.

Unfortunately by the time the value of community is lost enough that it affects the bottom line, it’s too late. Even if the site survives, a lot of the vibrancy is gone. For all we know, it is too late and there's nothing that we can do. I'd rather think otherwise, but I could be wrong, and as a responsible person working with, and allegedly leading a community, I need to sometimes re-examine if it is worth the effort to try to save things.

9
  • 11
    A moderator must have some activity every six months or they will be removed. Maybe we wouldn't be here if the same thing was required of those higher up. They wouldn't even have to do any flag handling — seeing a Stack Overflow mod at work would probably be enough.
    – Laurel
    Jun 11, 2023 at 5:05
  • 6
    I don't really think that's 'enough' - but rather someone, somewhere in the decision making chain has to be truly and organically involved in the community. Jun 11, 2023 at 10:18
  • Thanks for this answer. You've explained it well; I understand, and agree that this must be at the root of the problem, this and not listening (not having to agree, but to really listen, with a desire to understand) those who have been and are "in the community". Jun 11, 2023 at 11:49
  • @Laurel - Term limits for mods would also be good. I see a lot of great users with loads of energy who'll never get the chance to lead their communities
    – Richard
    Jun 12, 2023 at 11:49
  • @Richard That's a completely different issue, which you could discuss under the relevant meta question.
    – Laurel
    Jun 12, 2023 at 12:11
  • 1
    @Laurel - It's connected because by "community" you mean "those few individuals lucky enough to become community mods"
    – Richard
    Jun 12, 2023 at 12:16
  • 4
    I work with quite a lot of community who are not moderators, quite a few of whom are adversely affected by the decisions made. And I ended up a moderator because I was active in the community and folks felt I'd do good as one. I'd also note a lot of work I do around the community doesn't need a diamond, especially as I ended up transitioning more to and to a large extent, its just a way to speed up the 'janitorial' side of things. While I do realise 'quitting' is hard, I don't see why an active moderator needs to quit cause 'they've been doing it too long'. On the other hand, calling up Jun 12, 2023 at 15:15
  • another moderator slot is trivial, especially if there's community growth. We just ask Jun 12, 2023 at 15:15
  • 3
    For years SE has undervalued the community (which is just as valuable than huge curated database of knowledge they want to use to train AI). I've said before that healthy human-scale communities solve many of the issues SE faces. Attracting users who don't engage as community members isn't sustainable. They can get answers anywhere (sometimes scraped a republished from here). They can't get human connections from a search engine.
    – ColleenV
    Jun 12, 2023 at 18:47
31

Disclaimers: I have never worked for Stack Exchange in particular, nor do I have access to the Teacher's Lounge, so the following is based on public information and my subjective experience working in a couple of other large corporate tech environments. I cannot and do not speak for anyone other than myself. I would also like to emphasize that this answer is intended as an explanation rather than as an excuse. I do not approve of how SE, Inc. has handled this situation and nothing in this answer should be taken as a defense of their actions.

By far the single most common error I see when people talk about tech companies is thinking of a tech company as a monolithic entity. This is simply not accurate. It comes closest to being right when we're talking about very small startups that have not accepted VC money, but frankly, that's rare at best in this industry.

Quoting myself from an LWN comment:

A large corporation is difficult to steer from the top (because of the sheer number of decisions to be made overall) and impossible to steer from the bottom (because the most you can do is convince one manager at a time to support your way of thinking). The people who are really steering are hundreds or thousands of middle managers, each of whom has a slightly different understanding of the business's high-level needs and goals, a moderately different set of internal resources and levers of power, and a radically different understanding of the business's immediate problems. Every now and then, their collective decision-making is bad enough that an executive gets involved and tells them "you screwed up, now fix it." Every now and then, their collective decision-making is good enough that the corporation does something that looks very smart, to an outside observer. But most of the time, the corporation drunkenly shambles in the general direction of profitability, while the outside world tries to figure out what on Earth it could possibly be thinking. This is the wrong question. Corporations don't think. The people running the corporation think, but there are many of them and each one has a peculiar and limited understanding of the corporation's broader scope.

Of course, Stack Exchange, Inc. is too small for this model to apply exactly. But that doesn't mean we can completely ignore the problem. SE is still big enough to have significant chains of command, and is also owned by Prosus, which may not be directly on the org chart but will nevertheless exert influence over upper management. I tend to assume that whoever's in charge of the CM team (presumably Philippe?) is not reporting directly to Prashanth Chandrasekar (the CEO), so there's at least one layer of middle management in the way. We know from Jon Ericson's blog that management at SE was rather "feudal" when he worked there, so even one layer of middle management could impede communication significantly. It is also possible that the ChatGPT diktat was issued by someone other than Prashanth, and so that person needs to become involved in the conversation, which may imply another layer of middle management. The three (or more) of them are now (presumably) playing an elaborate and high-stakes game of telephone, because in a corporate context, it is considered highly irregular to "jump" the chain of command. I tend to imagine that, while Prashanth is probably aware that "the community is upset and the press is writing negative things about us," it is significantly less believable (to me) that Prashanth is directly involved in resolving the issue, because that would normally be the responsibility of Philippe et al., and Prashanth may not understand just how extraordinary and damaging the situation has become.

To make matters worse, community moderation is not a line item. That is, the current strike might not look like a loss, from upper management's perspective. There may be (speculating!) nothing on the books that explicitly says "The community is giving us spam filtering in the form of SmokeDetector and Charcoal, completely for free, and here's all the money we saved by not having to develop and maintain our own anti-spam solution," nor are there similar line items for any of the other things that elected moderators normally do. And so, when the community support is yanked, it doesn't look like the company just lost a valuable asset, even though it did. The accountants will tell you, wrongly, that the cost of the strike to the company is $0, because there is nothing in the company's ledger for them to add up. (To be clear, $0 is a correct response under traditional rules of accounting. It's just that traditional rules of accounting cannot possibly include all forms of economic loss.)

If true, that's frankly an indictment of SE's upper management. They really ought to understand where the economic value of their company comes from, regardless of what's in their ledger. It's possible, even likely, that my understanding here is badly incomplete. It may be the case that SE fully understands the cost of this, and is hoping to placate the community long enough to "pivot" to some other form of content generation (i.e. they want to replace us with LLMs, and keep us quiet until they figure out how to do that). I'm doubtful that would actually work, and as a rule I am skeptical of conspiracy theories involving tech companies, but the near total lack of communication from the company makes it difficult to completely rule out as a hypothesis. In my opinion, the more believable explanation is that the CM team (not SE as a whole, just the CM team) is trying to placate the community in order to survive until next quarter, not in order to eliminate community contributions.

This is unsustainable in the long run. Eventually, the company will realize it has to develop a spam filter, put it on the agenda, get it budgeted and approved, etc., and I would like to think that somebody will ask "why didn't we have to develop that earlier?" Eventually, Philippe will communicate to upper management that this is a "real" crisis, and not just a bunch of hot air in the press. Eventually, eventually, eventually, it'll all percolate up to someone with the ability to rescind the ChatGPT diktat, and that person will hopefully(!) realize that this situation is threatening the company's long-term viability. But it is hardly surprising to me that it has taken them more than a week to do that.

12
  • In other words, if we, the community, want to communicate directly with the CEO, we should be able to estimate our "value" (how much money we are saving the company). Right?
    – Rubén
    Jun 10, 2023 at 22:30
  • 9
    @Rubén-PeopleFirst: No, it's worse than that. My hypothesis is that we can't communicate directly with the CEO, because he has delegated that responsibility to Philippe. The money is just icing on the cake.
    – Kevin
    Jun 10, 2023 at 22:31
  • I wonder if we might find a proper community speaker to do the "elevator speech" thing IRL.
    – Rubén
    Jun 10, 2023 at 22:36
  • 13
    I agree with a lot of this, but I find it incredibly frustrating nonetheless. Not just because of my level of respect for their intentions (eh?), but also, and mainly, because it's such a wasteful and ineffective path. There is such a huge potential here over the next few years for profit. Not only without violating the social compact with the community, but in fact drawing upon it. Pretty soon, nearly every source of human language interactions is going to mixed with generated posts. Training on that contaminated data is... problematic. But they're seemingly set on killing the golden goose.
    – BryKKan
    Jun 10, 2023 at 22:37
  • 2
    @BryKKan they might have decided that what they have already (decade(s) worth of content), is enough for training?
    – Levente
    Jun 10, 2023 at 23:04
  • 3
    @Levente I wouldn't know, but if so that's almost worse to me, since that's demonstrably impossible today, and quite easily shown. Even just in tech, there are always new versions, new major programs, and even entire new languages. And these AI generators suck at emulating useful expertise without a direct example, or we wouldn't be having this conversation. In order for tools like ChatGPT to remain relevant, they must constantly be refined and extended, for which they need a reliable, and reliably human, live data source.
    – BryKKan
    Jun 11, 2023 at 4:56
  • 1
    Thank you for this answer and the link to Jon's blog. I had to chuckle at this line about SE Inc.: "...management has discovered a way to spin gold into straw." Jun 11, 2023 at 18:00
  • good post but I take small issue with your quoted comment... corps do think, just in a way so alien to us that we can't relate. Think, employees are ants and the corp is the ant colony. Neither can truly understand the other.
    – AakashM
    Jun 12, 2023 at 8:26
  • 3
    Quora now has a CHATGPT bot embedded into every single answer page, above the human-written answers, so it's not at all far-fetched to imagine the leaders of SE doing that
    – Richard
    Jun 12, 2023 at 11:53
  • 5
    @Richard That seems to me to be an improvement over intermixing the bot answers with the human ones. At least that way, the people who want the human answers can focus on the human answers. Jun 12, 2023 at 17:53
  • 1
    Yes, in the short term Stack Overflow (the web site) will not be affected much (there are already 23,748,099 questions (way too many, BTW), and a little more noise will not have much effect). But it will when most users lose confidence that there is any value in Stack Overflow (search engines start to only return questions with ChatGPT-generated answers and 90% of the answers to new questions are robotic (figuratively)). It doesn't matter if there is good content in there if the search engines don't expose it. Jun 13, 2023 at 16:03
  • @This_is_NOT_a_forum: That's not the point. You know that this is bad for SO, and I know that too. The point is that the company doesn't "know" that (insofar as it's not a line item anywhere on their ledger).
    – Kevin
    Jun 13, 2023 at 16:11
28
+50

The profit motive is simply too corrupting.

And there is a genuine alternative. The root of the problem here is expecting corporations to be nice people. Yes, with enough isolation from shareholder wrath, a "not a 501(c)(3) nonprofit" can do good for awhile... until something happens. The owner or charismatic, successful CEO gets hit by a bus. They get tired of being Mr. Nice Guy. They get into a money jam and VC has other ideas. Or, like one company with their passion for consumer privacy, that's part of their brand identity. Or they are simply cowed into doing the right thing because they accurately understand social media backlash.

But past habits or "brand identity" is not legally enforceable in any conceivable way. What is enforceable is their obligation to shareholders, and the universal solvent of that obligation is money. That's how (non-nonprofit) companies are defined by tax law, and that's that.

"But Harper! What about Firefox? Wikipedia?"

Non-profits (not-for-dividend) are of course the alternative.

Code.org and EFF are other examples. You can find out if a company is a nonprofit by visiting http://www.guidestar.org/ and typing in their name. At that point you can read their Form 990 tax filings, which state their activities.

With nonprofits, tax law requires a charitable mission and then unhooks the company from the profit motive. The charitable mission is a complication - there needs to be a plausibly charitable purpose, so you won't see a lot of charitable online brokerages or eBay competitors. But building a repository of knowledge, or giving people free help, are definitely not a problem.

That said, we see time and again organizes who are eligible for a nonprofit status choose for-profit - from not knowing any better, or finding it hard to change, or looking for that pot of gold at the end of the rainbow. In the last few years TechShop's founders lamented not organizing or transitioning to nonprofit status. Techshop only grew by partnerships, but they had precious few takers - especially among what should've been their mainstay, schools - because of the conflicts of the profit motive. Schools started their own maker spaces instead of leverage Techshop's experience and network. (e.g. allowing members in one unit to work at any). And of course they couldn't bolster their staff with volunteers. Who wants to work for free to line someone else's pocket?

This is the basic untenability of Stack Exchange's position here. There is always going to be this sliver of mistrust, and things like this and Monica just open the sliver into a chasm.

Existential threats don't bring out ones best

The data Stack Exchange management has to work with here is a mess. ChatGPT's release has been an inflection point for the site. Obviously there are fewer questions as some people ask ChatGPT. But read their post (I believe their data)... they're also seeing a notable (2.2% per week) decline in active (3 answer per week) answerers, simultaneous to moderator bans rising twentyfold from 0.4% of active answerers per week to over 7%. Any pointy-haired boss can plainly see: Where did those 2.2% of precious long-term contributors go? They're being suspended by overzealous moderators trying to stamp out the ChatGPT forest fire! See? I can do science! Yes, yes, all that "correlation is not causality" stuff - philosophy is all well and good, but not when your Pot Of Gold is at stake!

See what I mean about the profit motive?

And in all my years in nonprofits, I've never met an Executive Director who didn't have the seed in the back of mind - "How can I replace these pesky, demanding, troublesome, uncommandable volunteers with something else?"

So here we are. Sort of like the Russians, when you start failing, teamwork is the first casualty and everybody points fingers and snark at everybody else, and assumes the worst of each other. Stack Exchange can't decide whether ChatGPT is public enemy #1 or our replacement.

Where does that put us? I'm reminded of the tale of The Scorpion and the Frog, where the frog does volunteer labor on the scorpion's promise of being nice. The promise is even reasonable. It would be foolhardy for the scorpion to violate it. And yet, the scorpion does, because that's what scorpions do. And that's also what for-profits do when their pot of gold is at stake. These world changes are an existential threat for Stack Exchange management and owners, and that makes them unpredictable. "Nice" is the last thing on their minds.

Anyway, now you know that scorpions do sting, and if you don't want to be stung, there are platforms that are not scorpions, and have a higher calling than "profit".

2
  • 1
    Yes, it's interesting times we live in. If moderators are right and can convince the company, then they would even save the company by this strike. But as it is, it's probably a long way to convince them of that - especially because they seem to believe the opposite to be true currently. On the other hand as anongoodnurse here states, we don't strictly need them. So they better listen and start making better decisions. And if not (say wait like 2-3 weeks but not much more) start leaving, I'd say. Jun 12, 2023 at 7:26
  • 3
    @Trilarion Well if we creators decide we don't need them, I certainly hope we choose our next platform to be more resistant to such ... influences. Jun 12, 2023 at 18:30
21

I happened to watch this recently, maybe it's relevent:

It includes:

... superior management skills. But it should go without saying that a private equity company with little or no experience from a particular industry cannot improve the actual products or actual services of a target company. All this private equity partner, this glorified financial intermediary, can do for its target company is to deploy the bluntest instruments that capitalism has to offer. The private equity firm helps the target company cut costs through layoffs, offshoring, and asset stripping. And the private equity firm helps the target company increase revenues by deploying state of the art methods for price increases. In other words what private equity firms do is to help target companies focus more on financial engineering and consumer exploitation and less on improving the company's products and services.

I guess it's up to you, all of us, to decide whether contributing to these sites/communities is still suitable a vehicle/medium for you.

FYI I also noticed that the new owners seem to categorise Stack Exchange as EdTech (Education Technology).

My mum was a preschool teacher -- her salary was lower than mine, and I thought the reason for that is obvious, i.e. not that her work was less skilled but that she was funded from a fraction of the school fees paid by 20 or 30 families per years, which doesn't "scale" very well -- compare that to rock stars and movie stars, who sell to millions of people, or even engineers.

Perhaps it's inevitable that "financial engineers" who try to make EdTech more profitable will look to replace or augment the human labour and develop expertise in AI.

Yes a lot of things about it are imperfect; I suppose they see it as a work in progress.

10
  • Thank you for this answer and the link. I'm not business (or tech) savvy, so I'll learn from this. Not being tech savvy, and forgive me my (extreme) naivety, but is AI seriously expected to adequately replace human expertise? The only AI I've followed closely is IBM's Watson, because it was supposed to improve medical care, which it failed at dismally. I'm sure AI can do a lot of things better than I can (my calculator can, too). But replace human expertise? That's... complicated. Jun 10, 2023 at 19:23
  • 6
    @anongoodnurse: Speaking as someone with a little knowledge of the field (I am not an expert), IMHO this is a complicated and difficult question to answer. Historically, we have heard "this AI will change everything" many, many times before, but in practice, the resulting AIs were almost always very "narrow" constructs that struggled to do much of anything outside of some highly specialized context. For example, Stockfish plays very good chess, but it can't do anything else. So my default position on that is one of deep skepticism. But OTOH, new technologies are new. 1/2
    – Kevin
    Jun 10, 2023 at 19:50
  • 2
    LLMs and other generative models are seemingly applicable to a wider variety of use cases than prior approaches. They still have a lot of problems, and it's not obvious that those problems are solvable, but there are a lot of very smart people who are at least trying. In the past, every AI summer has been followed by an AI winter, but no one can say with certainty that that will happen again. 2/2
    – Kevin
    Jun 10, 2023 at 19:50
  • @anongoodnurse Well I like how well Google guesses what I ask it.. But I barely know AI theory, I never studied it, and I don't know the state-of-the-art. They say it's getting better than some years ago? I believe that AI is already used in some medical domains: for research; to assist people reading images; and other topics, which I cannot or should not talk about, but it's already a reality to some extent. Maybe in medicine it's a tool to assist doctors and nurses, not an independent replacement. But e.g. autonomous vehicles exist now too, evidently some "human" expertise is automated.
    – ChrisW
    Jun 10, 2023 at 19:51
  • 2
    @anongoodnurse "but is AI seriously expected to adequately replace human expertise" — I recommend Tim Urban's articles, from 2015 (!) I find these incredibly eye-opening contributions: part 1, and part 2. They are long, but really worth reading, because this topic will only get more and more relevant.
    – Levente
    Jun 10, 2023 at 19:57
  • 1
    @ChrisW - I agree it has helped with interpretation of images, but not much of anything else (I don't do research anymore so don't know about that.) EMRs were supposed to assist doctors, nurses, and patients, but the only thing they have done is given many health care practitioners the final straw needed to leave the profession. I am quite serious. Nothing they do has helped us or our patients; it has made insurers richer, and has made HCPs slaves to the EMR. Robotics, on the other hand... Yes! Jun 10, 2023 at 20:12
  • 1
    @anongoodnurse google.com/search?q=ai+medical+alarms suggests they've been developing algorithms for bed-side monitoring -- whether and when to generate an alarm when physiological signals changes. And I think that might have been productized (now in service). The term "AI" is very broad and also a moving target, an old science fiction claim was "Any sufficiently advanced technology is indistinguishable from magic", for me personally the term "AI "is more-or-less like that. The kind we're talking about here is presumably natural-language (i.e. human language) processing, using ....
    – ChrisW
    Jun 10, 2023 at 20:25
  • ... so-called LLMs, which as I said I know nothing about. :-) One tool I've noticed in recent years is DeepL which IMO does a remarkably good job of translating between English and French. I don't know but imagine that too is trained on a corpus of human-translated texts, e.g. the publications of the Canadian government. Maybe that i.e. machine-translation is an example of an application where the machine can seem to understand human language, or at least where its output is good enough to help humans in some way.
    – ChrisW
    Jun 10, 2023 at 20:30
  • 1
    @ChrisW - This is off-topic so I won't say more, but one of my sons, a nurse, would disagree with you most vehemently. Responding to the now voluminous false alarms are taking up an enormous amount of time better spent actually nursing patients. Another example of the tool making HCWs its slaves. Jun 10, 2023 at 20:31
  • 2
    @anongoodnurse If you clicked on the link I posted you might see that's the problem they've been trying to solve, i.e. "alarm fatigue" caused by false alarms -- so they'd like to get better at deciding whether an alarm is warranted etc.
    – ChrisW
    Jun 10, 2023 at 20:34
12

Without us, SE would not be what it is. It would just be another Q/A site with low SNR and no community. That said, I am not sure SE Inc needs us anymore. The Q/A part of SE hasn't been their business model for a long time. The jobs and teams products don't really need us. Selling our past content and votes to train AIs doesn't need us anymore. I guess without knowing what the SE Inc plan is to make money, it is hard for me to see why they need.

As for us not needing SE, I think that is wrong too. What we built here, with the help of the original SE team, is really special. There really are not any alternative Q/A sites with a high SNR and a community for those who want that.

3
  • 4
    They ditched the jobs product a little while back. Might still be offering some targeted advertising for jobs but no longer a specific job search/matching program. Jun 13, 2023 at 2:32
  • 4
    If they want to shut down the site and only sell pre-ChatGPT data, that is going to be a less and less attractive offering as the content ages. If they want to keep the site running and keep the quality even modestly close to what was possible before the strike, they need the community to exist, or a radically different model for quality assurance.
    – tripleee
    Jun 13, 2023 at 5:04
  • 3
    @tripleee I think their analysis of the chatgpt data suggest they are focusing on the easy to quantify number of users who provide 3+ answers a week and not on the more difficult to quantify quality of those answers. As I said, we don't know their plan, so it is hard to say if we are needed.
    – StrongBad
    Jun 13, 2023 at 12:31
10

What will it take for me to finally leave the network?

Successfully breaking bad habits usually involves replacing them with good ones. For me anyway, I think it will involve the emergence of a better option. Bringing this to fruition may involve some additional up front work but hopefully the long term benefit will be something along the lines of what we imagine SE should be.

Are you active in other communities? Are there other networks you think may be suitable for a Q&A site serving your field, which doesn't yet exist?

4
  • 3
    I'm glad Codidact is there. I suspect, though, what I need to do is to start contributing to a cause and/or community in real life. Jun 13, 2023 at 10:12
  • 1
    What's different from the previous corporate ****storm in 2019 and the present one, is that the previous one is what caused Codidact to emerge, so by now it's a relatively mature platform. Most of the quirks and bugs have been fixed, what it needs the most is just more users.
    – Lundin
    Jun 13, 2023 at 11:51
  • @anongoodnurse: It is always an option to (literally) remove litter from the streets of wherever you live. It doesn't require any organisation. Start doing it today. It doesn't matter if you subscribe to the broken windows theory or not. Jun 13, 2023 at 15:29
  • @This_is_NOT_a_forum - How uncanny... One of the things I was contemplating doing was picking up trash! Picking up trash seemed a quick and easy contribution, as it is everywhere. Maybe I'll sew up a bunch of cotton bags and hand them out at the grocery store since I'm so anti-plastic. I'd like to help people, though, and I can't do that medically any more (retired my license). Jun 13, 2023 at 18:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .