« first day (5161 days earlier)      last day (14 days later) » 
05:00 - 16:0016:00 - 00:00

5:32 AM
8 messages moved to Chimney
 
2 hours later…
7:53 AM
@SPArcheon-onstrike I think a lot of us want this place around. Despite the best efforts of some of the people who run the place :D
8:15 AM
@JourneymanGeek I bet than even more don't want to be played as idiots any longer.
And trust me, I am trying to NOT be too snarky.
Otherwise I would have asked on that post if amidst this totally required top priority the network can't live without piece of trash move they found 5 minutes to ... TALK WITH THEIR LEGAL TEAM LIKE THEY PROMISED CELESTIA KNOWS HOW MANY MONTHS AGO
There's disagreeing with choices, there's even complaining they're taking too damned long to do things. I've pushed these things the 'right' way and will continue to. "Killing off smokey" and "let this place burn though? isn't right tho
Because you know, when you have serious PRIVACY COMPLIANCE and GDPR related concerns, finding new ways to scam and alienate your users is top priority
@JourneymanGeek Good for you that you can be happy with continuing to be played as a fool.
Just let me know... you pushed those thing, you got promises for answers... did they come?
I didn't see them, so if this is the usual "private stuff can't comment" idiocy let me know
I'll keep pushing. Till something happens, there's something to share, or there's nothing left?
Well, that's a problem innit? the community team's easier to get a reach of in private spaces
but I can't share without there being something public
Well, part of an answer but nothing concrete
@JourneymanGeek so, when an employer abuses its workforce, should the employee continue sending mails to the trash compactor or should they raise the stake and go on strike again?
Because the company has clearly shown they don't understand any other form of language.
honestly if at this point it probably wouldn't should be striking, its walking out. but there's probably other potentially extencisial threats, some of which worry me more personally than this
8:24 AM
and even with the strike... just go back on the tavern and you will see that the general sentiment is that many of the striking mods back then were a little to fast to "we won, let's go back to work"..
I think for most mods the concern's their communities, more than the company
@JourneymanGeek I am starting to question if status quo doesn't come even before that...
8:38 AM
@SPArcheon-onstrike Or maybe people are more attached to the idea of SE than the reality
8:51 AM
@JourneymanGeek The Akasha was LESS corrupted.
Alas, forgive me. I got back on my intent after reading this
> Since SE decided to respond to everyone except me
they deserve the cynical answer.
And yeah, those things are issues
there's also other issues surrounding that that need addressing
> A letter to Santa should be placed into an envelope with a postage stamp, and sent to 123 Elf Road, North Pole, 88888. Don't forget your return address — be sure to write it clearly and include an apartment number if applicable, so there is no confusion about where the gift should be sent.
They can be addressed there ;)
@ShadowWizard it's not an own answer -- the question is from Slate, the answer from Cesar
@balpha I think there's mention of the ownership moving tool being used
it was a self answer at some point
oh yeah that would explain Slate's avatar in your onebox
9:02 AM
@SPArcheon-onstrike not enough of an answer but something But apparently patience is something I have way too much of :D
@JourneymanGeek indeed, the owner was changed right before the answer was accepted
"If a tree falls in a forest and no one is around to hear it, does it make a sound?" is a philosophical thought experiment that raises questions regarding observation and perception. == History == While the origin of the phrase is sometimes mistakenly attributed to George Berkeley, there are no extant writings in which he discussed this question. The closest are the following two passages from Berkeley's A Treatise Concerning the Principles of Human Knowledge, published in 1710: But, say you, surely there is nothing easier than for me to imagine trees, for instance, in a park, or books existing...
or to put more bluntly... if that something is available only on non-public-board with the usual NDA catch 22 agreement, a private exchange between you and some lovecraftian pantheon entity you still can't disclose or a tree fallen in a distant forest where no one saw it....
Do I should care for thing that I can't know, I can't even tell if existing or what?
I'm not sure
I'm still trying to work out what next between a lot of events
I mean, good for you if they privately reply to you.
I'd LOVE to be the bearer of good news for once
but as a moderator, and a regular user, I've a little more access but those places also have their own rules
9:08 AM
But... do they realize that there is a community out there that is not JourneymanTheGeekWePrivatellySpeakWith or a mod with access to the secrets of the Teacher Lounge?
Do you expect us to just assume that "something good is happening you just can't see it yet" when instead most of the time it is Zoe sadly crying out that ...
in Zoe's Meta Foxhole, Sep 5 at 16:45, by Zoe - Save the data dump
The shoe hasn't dropped yet
and she can't tell us what the shoe is because Stack obviously don't allow the mod to warn us of any trash move they are planning in adbance?
Nope
or how many shoes there are
Its not an ideal state of matters but there's only so much any of us can do - and ideally we try to ensure that any particularly harebrained idea we know of is... tempered a little if we know about it befire
Do you expect me to just "trust" that there is good faith in some exchanges I can't see when what I see is again Zoe telling me that
> Since SE decided to respond to everyone except me on the internal announcement,
And wonder what, it is pretty blatant that this scurvy move was made EXACTLY to kill projects like the one Zoe host.
IE, once again, desperate attempts to gatekeep the dump.
ah naw, that implies that we're important enough. This is very much around the whole AI obsession
Everything else is collateral damage
9:14 AM
Had I to guess ... based on that one post about a certain exchange a certain ex employee posted some time ago, I may have a pretty clear idea on how this "fix" came to be, who asked it and why
I don't even think the issues Zoe had were directly related, it was some cloudflare thing
@JourneymanGeek ... have you read all Zoe post?
Yup
I'm probably aware of a bit more of everything that's going on - including that than I let on
then you should see that my previous statements were quite specifically about the "Stack Exchange data dump downloader and transformer" which is "assumed" to be unaffected, but yet they avoid to answer about until this went public.
I know it works
I'm also still waiting on the promised complete dumps from SE
9:22 AM
So, first issue - I wonder what is the purpose of a private closed door discussion on a private board if posts like Zoe's are surgically ignore in what can only astern to ill intent (and we even get answer falling out of the blue like they never knew about Zoe concerns before).
Second issue, again... want to talk about some of the comments there?
@SPArcheon-onstrike I mean, on my part, I know community managers will respond to it, and hopefully move things along.
I specifically know that Zoe's issues are with the comment archiver tool breaking + that's CF related
That's one of the things a formal chat API would 'fix' in theory
(which I am also pushing for?)
> "I can tell you that browser emulation goes against the spirit and goal of these rate-limits, if nothing else"
Its the only way to get some things done I guess
we wouldn't need the data dump downloader if there was a requestable full data dump
This is Slate, just after Zoe specified that her tool works as a browser automation
Which I would add, is one of the things I am pushing on
I know how the tool works ;)
9:26 AM
so, now, please, insult my intelligence again then. Tell me again this [redacted] move with this precise timing is not about killing down tools like Zoe's.
It was planned ahead of time, and isn't about specifically killing tools like Zoe's
A tool whose nature "goes against the spirit and goal of these rate-limits, if nothing else"
A tool entirely un-necessary if not for the data dumps not getting altered in the current way
also the tool automates the process of clicking on each download, with a valid account and downloading the files...
its nearly impossible to block
I don't see what this line would want to add. Are you aware that this claim actually just proves what I have just said? They have their agenda to put some walls and fire ring hopping around the dump, it is clear the entire community hate it and it is clear that they don't give a [redacted] since it is working to THEIR goal.
So, exactly... why should I don't think they are trying to trash alternatives too?
Her chat archive tool though, folks find handy, and ironically the part that's broken is not the data scraping, its the data archiving to resources SE controls
@SPArcheon-onstrike Imagine if you're a little clueless and are worried say openAI does the same thing, instead of paying them.
9:34 AM
@JourneymanGeek imagine you are an executive that thinks SE can track who download a file from a third party archive site and also imagine you have a deal with some AI company for premium access to your tasty knowledge archive, how would you try to prevent third parties from accessing the same data for free when said data is infinitely freely reproducible because those "meddling kids" that sold you the site made it CC?
I mean, I'm smart enough to know you can't stop the signal
I'd ensure the intent of the data dumps are met, in a better way than what they did
If this wasn't clear enough, after THAT leak, I don't expect the forced hand changes Stack is going thru to come from someone who has a technical knowledge of how things work
SE wasn't always on the internet archive, and I'm pretty sure that there'd be folks willing to have the data in escrow
Nope
@SPArcheon-onstrike technical, political, social...
to borrow a term I once heard someone said - you'd probably need someone who could part the waters and pull off at least a minor miracle
(or for the AI bubble to burst sooner rather than later)
The scenario in my mind is more akin to some CEO getting up in the morning and being scared because they saw on "www.fakenews.com" that you can get Covid from https and so imposing to some unlucky dev to remove https from the site, no matter how much said dev tries to explain this is stupid.
 
2 hours later…
11:46 AM
@Catija ah, I see. Thanks!
@EleezatheOtherWorldWizard yay! Awesome, and waiting to hear it. :D
@balpha was originally posted by Slate as well. ;)
@Catija I'd actually forgotten that we can reassign ownership. So I can do it. I just need to wait until tomorrow to ask at work for a willing <s>inbox notification victim</s> volunteer to graciously take on this answer's ownership. — Slate ♦ 14 hours ago
@SPArcheon-onstrike or personal revenge against a mod that annoyed them too much?
12:06 PM
@SPArcheon-onstrike No.
@JourneymanGeek You really should realize that nothing they tell you in private can possibly be valid as long as they refuse to publicly say it. It’s basic reasoning.
@AndreasmovedtoCodidact if you have better ideas than engaging in good faith, you should try them
@JourneymanGeek Yes. Go eat ice cream.
That’s delicious.
12:53 PM
@JourneymanGeek Depends. None of the "solutions" to get things moving (non necessary in the right direction nor to even reach an actual closure) that I can think of would NOT be very destructive.
I am past the point of thinking you will be able to get anywhere by talking.
Just look at you.
You started talking MONTHS ago. You haven't even resolved the issues you raised back then, and you are kept hostage from talking about the empty exchanges you had.
In the meantime they were able to sleep on the issue, MAKE NEW ONES and getting worse... and worse... and worse....
All while you wait, sure that SOMEDAY they will address the first problem
Again, call me cynical... but this feels very convenient.
Let me get my Nahida wig...
1:07 PM
months?
Years
BUT the alternative for me isn't getting mad, its walking away
1:25 PM
@balpha nothing in our logs, I'm starting to suspect Cloudflare...
Long ago in a far far away land, a white dog lived in harmony with their friend working at the royal library. It was a simple and quiet life, working together to record all the knowledge in the land in books and scroll. And the dog was happy with their carefree life, scribbling scroll and revising old manuscripts all day.
But all things come to end and a sad day the king died. The throne went to some unknown lord from a land clad in winter, someone they never heard of before.
Sad was the life of the poor animals at the library as they slowly realized the new king had no interest in curating
@ArtOfCode I guess that serves us right 😬
@JourneymanGeek or if you prefer... this
... don't make the mistake to think the stories were useful and on purpose. As you can see the plan was to distract them, do nothing and run.
1:42 PM
@ArtOfCode here's what Josh says
oh there he is, hi @JoshZhang
o/
Funny story, I'm going to be writing a meta wiki on Cloudflare, what it does and why.
erf
@JoshZhang there seems to be a bunch of friction due to ... somewhat potentially overzealous cloudflare blocks. There's a particularly annoyed mod whose tools are getting mysteriously broken ._.
there's a delicious irony to the fact that we are getting blocked by cloudflare, too 😬
@balpha Possibly but you always get blamed...
I'm aware and actively making adjustments in the background. The problem is sometimes it's a moving target. Cloudflare prefers all web page traffic be human because there are a lot of malicious actors on the internet right now. We can't turn Cloudflare off because that leaves us exposed.
1:57 PM
The good thing for us because we're an enterprise customer is that there are more knobs to turn. Not so much on free tier.
Just wait and see when CloudFlare changes policy to automatically ending subscriptions for servers that now get blocked by interacting with other servers...
But CF uses the free tier to collect traffic patterns / data to help them defend against big attacks.
They're not some mean company that hates scripts, they actively support an open internet, partially why they offer the free tier to everyone. The other benefit is like I said above, they get lots of good traffic data to help defend against large scale attacks.

Long term I see all CDN's saying automated traffic needs to be directed to API's while normal pages are for human behind browsers.
@balpha Actually, the alternative would be worse. The CloudFlare monopoly, only accepting interactions from other CloudFlare users.
to phrase it carefully, I'm not sure how far away we are from that reality tbh
2:01 PM
@JoshZhang Ideally, I suspect people who write scripts would prefer that. there's just a lot of stuff that isn't, and I suspect the resources needed to do that are uh... either short or elsewhere.
@balpha somewere between "soon" "later" and "never" ? :D
I agree, our API doesn't cover everything currently. The issue for CDN's is that it's becoming harder to protect against malicious traffic and they need to steer everyone to doing the right thing.
It's more the fault of the bad actors but can't do much about them.
DDoS aside think about all those scalper bots, etc.
CF says "Definite Bot" >.<
@AndreasmovedtoCodidact just wait till they get some users randomly locked out because the old Winter Bash app (that I bet almost no one removed from the app page) is still doing some weird api call
@ArtOfCode well CF is not wrong ...
the user agent should be "SE Chat"
@SPArcheon-onstrike probably worth focusing on actual problems than deep hypotheticals :D
2:07 PM
@balpha Yep, I've just exempted the UA from the bot rules. Should be fixed.
nice let's give it a try
in The Nineteenth Byte on The Stack Exchange Network Chat, 10 secs ago, by Feeds
posted on July 05, 2024 by trichoplax‭

Given 3 points, output one that is separated from its nearest neighbour by the largest distance. Input 3 distinct points (that is, no 2 points are in the same position)....

bingo!
@JoshZhang The problem is that with an API, though, the control and responsibility is moved entirely to the website owner. So it prevents good actors from rightfully reading parts of a webpage intentionally or unintentionally not available from an API.
@AndreasmovedtoCodidact Ideally, folks here should be savvy enough to request what's missing
@JourneymanGeek who said I was focusing on anything? Do I look like I am worried if some unneeded change only made because of a secondary agenda breaks the site? I was just pointing out that if something broke due to poor clean up of old features that were killed because they made no profit... I would laugh hard.
2:11 PM
@JourneymanGeek No, also, I'm not talking about just here.
Sadly, I think this is a this is why we can't have nice things situation. Bad actors are causing enough "damage" to make people need to clamp down.
@JoshZhang on the other hand - there's unfortunately a lot of collateral damage in the process, both in terms of resources, and good actors getting disadvantaged
Whenever we put anything in at least on the SENetwork side one of the first things we ask is how will this affect users. A few years ago when we got DDoSed hard and we needed to temporarily block Tor I said in the call that people would be rightfully upset.
It's a really tough balancing act right now.
@JoshZhang If this is about the recent change, I will have to point out that many users asked and are still waiting for a disclosure of the actual reasons behind those needs, so that we may be proven wrong into assuming this is the next step in protecting the data OpenAI bought and probably would love have exclusive access to.
just look at this one for example
So this post says what SE is doing, but what I can't figure out is why this is being done. What problem are you solving? Are there too many search engines that overload your servers? Or are you very upset by language models learning on the Q&A content? Or what? — Ruslan 5 hours ago
or this
46
A: Preventing unauthorized automated access to the network

Franck Dernoncourt We need your help Many of us don't want to help you sell our content, which we agreed was under CC BY-SA, i.e. allowing commercial use. You're killing the data dump (which already was incomplete, e.g. no images), and now you're killing scraping. The original spirit of Stack Exchange was to shar...

Nothing against you personally, but let me say that "we can't have nice thing because people evil outside" comes out quite empty when we are not allowed to know what exactly these evil people are doing.
I can't give the you the official reason why for obvious reasons but I can tell you one of the main benefits I'm looking for from the change is reducing erroneous traffic that will cost the company a ton of money. Requests cost money and compute, we've had a steady uptick in traffic but a ton of it is from bots. The web site is for humans.
2:19 PM
Since you mentioned scalper bots, that's a good argument in favour of not putting up a blanket-ban on bots. The best option, considering the goal is to allow good actors access, is to strangle just what those bots need, and not the rest. Reading content on a page doesn't need the same guards as creating accounts or using those accounts for something specific. For DDoS, do you really need a blanket ban? Is it not possible to measure when a DDoS attack is incoming, and only then defend?
Also is there any evidence that those bot existed before the need to scrap the site was crafted by the data dump changes?
@SPArcheon-onstrike there was a long standing request to deal with those
I've been here 4 and a half years and I've always been focused on the public site (as opposed to the Teams product). I've personally seen the traffic trends and I can say 100% bot traffic has steadily risen every year since 2020.
@JourneymanGeek let me guess. The scrapping mirror sites?
@SPArcheon-onstrike yup
2:23 PM
@JourneymanGeek The ones "we can't do anything about"?
so, suddenly they can after the OpenAI partnership?
@SPArcheon-onstrike Well, something is done no?
@SPArcheon-onstrike ah, well, politics. Someone managed to convince the higher up these things are needed
@JourneymanGeek Someone being OpenAI I would say.
I've no visibility on that ^^
"I'll pay you 20% cooler more if you kill off our competitors"
That's not what's happening.
I can say that at least lol
2:26 PM
@JoshZhang There is an official reason, but you cannot share it?
I'm not the right person to share it.
@JoshZhang I'll still be happier if this AI bubble burst without too much damage to the network ._.
@JoshZhang Unfortunately, it seems nobody is.
@JourneymanGeek I agree with you there
Also its worth remembering no one has 'full' visibility of what's happening
2:28 PM
There are a lot of factors at play plus we're owned by a publicly traded company so that also limits what can be said.
and to an extent, as much as you complain about it, information is siloed
Eh, I have a reasonably complete picture of what's happening. We want to be as transparent as possible.
Key words as possible =\
@JoshZhang I'm going to disagree a little - stuff comes out at different times to different people
Slate has been doing a great job of keeping y'all's concerns front and center
(in general, I'm juggling things, annoying people a lot the past two weeks :D )
2:32 PM
Thanks, noted / edited.
@JoshZhang sadly I can at least hope you realize that thanks to all the "We can't say" that is how it looks like for us. I mean, what would YOU think had you just our info? The company tries to gatekeep the dump, I (and many others) said something along the lines of "I hope you realize RPA exist" and wonder what that is the next thing you are working on after telling us for years that you couldn't do anything against scraping sites. How does this look to you?
@SPArcheon-onstrike that the current corporate obsession touches on the main sites, and we both need to try to mitigate the dumbest ideas, and get the company to do better on the right ones? :D
@JourneymanGeek I get your (reinforced over and over) sentiment, but how does this even reply to my question.
@JoshZhang looks like you (the company) put all the eggs in one basket, let's call the basket Cloudflare. Is that risky? Suppose they get hacked... SE might get a fatal blow. I don't remember what it used to be before, but isn't it better to have some alternative ready for use as well?
@ShadowWizard in theory they can and have switched CDNs before
2:37 PM
But looks like Cloudflare is more than just CDN, for SE.
they went from cloudflare to fastly to cloudflare. There's some engineering lift but its doable
@ShadowWizard it's a piece of infrastructure, wouldn't it be the same if Azure was hacked, etc?
But yes the company was originally on Cloudflare, then Fastly, now Cloudflare again.
@JoshZhang tbf, I'm sort of a fan of the old ways of "we run everything ourselves, with insane efficiency" :D
@JourneymanGeek probably not feasible anymore with the volumes of visitors.
but also at "insane" cost
We're simply not big enough to justify having a ton of hardware engineers keep a datacenter running / updated
2:39 PM
@JourneymanGeek I am explicitly asking how one can be expected to not assume the worst when multiple, easy to connect dots are painted in a short time frame, connecting the dots paint a upside down star in a circle with an horned face in the middle and all the explanations you get is "sorry, that is secret for now"
@JoshZhang I think Nick once offhandedly claimed you could run the network on a raspberry pi
You can, and it'll serve a whole 60 requests a minute lol
It's never been quite that lean. Maybe during the first couple years, or on v1
@JourneymanGeek run it on a PS 5 pro then, that is just 699$ for the digital only version.
The web front end is scalable but it still has to have a big DB back end
2:40 PM
@JoshZhang well, sadly yes, nothing is 100% safe from being hacked.
@SPArcheon-onstrike Cause I assume good intent a lot of the time, just not that we're on the same page. I'd also point out I do think many of the company's actions hurt the network, and have generally been vocal about these things
@JourneymanGeek you kinda missed the point again. The question wasn't for you. It was for the company. Or for the ones in this room now specifically.
Currently in an incident retro so the answers that need more thought will be delayed. I'm not dodging per se
I just wanted to see for once if there is a realization of just how much burnt out the community is and that the company CAN'T reasonably expect we to constantly assume "good intent" about changes that aren't public motivated and explained because "secret"
@SPArcheon-onstrike pro tio: trying to convey this sentiment is an exercise in futility. At a point, it's time to lean back and watch it burn to the ground
2:47 PM
1 hour ago, by Josh Zhang
Funny story, I'm going to be writing a meta wiki on Cloudflare, what it does and why.
If you want the honest truth, it's this: no, we can't really stop people from assuming the worst. What we can do is attempt to be transparent and honest to the fullest extent we are able to do so. Within those limitations, we can control what we say, but we can't really control the conclusions community members draw from it.
3
@JoshZhang Thanks, that will be much appreciated.
For example, I can tell you that to the best of my knowledge this change set is not motivated at all by attempting to restrict data dump access. But if you don't believe that's true even when I tell you, or you believe I've been misled too, then... I can't really force you to accept it as truth. In this sense, I recognize the burnout, and I respect it. All I'd want you to know is that I empathize, and I don't personally take offense because of it.
@Slate looks like those limitations are very, very strict, so even if you want to be transparent, you can't. Not even about the nature of those "limitations". But meh, that's nothing new, and I don't expect you'll suddenly start sharing something that you didn't before.
2:50 PM
@Slate Problem is that even the most well known and trusted of you folk there have a depressing history of being "removed" lately, so we can't even tell if you are free to speak your mind or you are in a nightmare fuel "don't be the next shog9" scenario.
@Slate That stance seems continously less common within the company, unfortunately.
@SPArcheon-onstrike I meant on an individual level. That MSE post was the last thing I needed to get done before being done communicating with SE employees, for example. There's no point in trying to communicate with an entity that has no interest in doing so
@ShadowWizard Partially, sure. In this Meta post I didn't really talk about the motivations, why we're doing this, what the grand vision is, and that's on purpose. There's a time and a place for that, too. But I have a dual responsibility. I am an internal advocate for the community, a role I am performing to the best of my ability (a part of my role I'll grant the community doesn't see day-to-day). But I am also responsible for helping the community manage change in the best way I know how.
In a sense, this Meta post is the latter. I picked a specific factual and technical tone, because my ultimate goal is to help folks manage change. If folks don't feel like they can extend that effort right now, I respect that. But as a CM, I have to try.
@Zoe-Savethedatadump In a perfect world scenario, you also wouldn't have needed to post that TWICE...
@SPArcheon-onstrike I'm painfully aware of that. There's also worse things about the situation that leads to the only obvious conclusion being that trying to communicate with a company that doesn't want to be communicated with is a waste of time and effort - time I won't be spending in the future
3:01 PM
This subject is a bit sensitive not in the sense that it's a company secret but more not tipping our hands to bad actors.

We've been getting more and more malicious traffic that has caused various undesirable things. Such as a ton of fake account creation and fraudulent ad click throughs. We are actively combating things like this and the Azure rate limiting is part of the mitigation strategy. It's difficult to broadcast the steps we're taking because it's not a passive "attack" any more. We see the other party at the other end actively adapt to the mitigations we put in place.
Is it the only reason we're putting the restrictions in place, no but it's one of the motivating factors.
define "fraudulent ad click throughs"?
as in "random people automating ads clicks"?
I can't get too specific as to tip off what I see but bots that normally don't trigger an ad impression now are at a scale where it's not acceptable.
3:19 PM
@JoshZhang just clarify the meaning, no need for full disclosure. You actually mean you see an increase in ads clicks that looks like it could be fraudulent? I was just confirming you meant "more entities clicking on ads on the site"
Correct, more ad clicks that are 100% fraudulent.
Man... quiet in here all week and then three pages of messages in one morning. Amazing what having staff in here does. ;)
Hey Cat, miss chatting with you =<
I am just struggling on the reasoning to perform ad fraud here.
Staff's job description: page out the chat.
3:22 PM
what is there to gain?
Correct me if I am wrong, but your ads provider is Google.
I don't know for sure but a few ideas:
- If your click through rate is garbage you can charge less per click or people won't advertise with you
- Disguise your traffic as human by showing you're clicking on the ads
- ???
- People suck
@JoshZhang Same! I do really appreciate that your no-nonsense way of putting things is still going strong. <3 CMs can do a lot but sometimes you just need someone who works with the code to put it more directly :P
GAM is one of our ad platforms
I doubt Google would engage in fraud to get paid more by their clients.
The source isn't Google lol
3:25 PM
@JoshZhang well, that would be quite stupid right?
And it's not a monolith like that. Google helps serve ads through GAM but the mechanisms on both ends are not always straight forward
But what I am thinking about here is "who is going to gain what"
I can't get any deeper for many reasons but one is I don't want to talk about customer technology
@JoshZhang Maybe a weird one but... would an ad buyer do it to boost the CTR of their ads? Like, the marketing team wants to show that the ads are a good ROI?
@SPArcheon-onstrike Who gains from DOS attacks on SE? Faking ad clicks is just another form of sabotage.
3:27 PM
@Catija possible but I'm not familiar enough with the actual economics of web ads to be able to make any inferences there.
@JoshZhang Neither am I. :D I would imagine a high CTR without an increase in sales probably wouldn't be of much value, anyway.
more clicks on your ads = you have to pay the ads provider more. But since the main provider here is Google I sincerely doubt they would want to do something so risky. So, while I don't trust Google on many other things (respecting GDPR and privacy being one) I also don't think they would scam their customers to get more money and risk a world level scandal.
So, I would put Google past my suspects list.
The companies paying for ads space? That would be idiotic for the same reasons
The provider isn't Google, they're the facilitator. Lets say my company AwesomeCo bought an ad through an agency. That agency buys inventory from us through Google (marketplace).
So there are more players involved than just the ad buyer, Google, and us.
@JoshZhang then that should be quite easy to track down to specific bad actors intermediary companies.
3:31 PM
How? Through the traffic?
I mean, I thought most of SO's ads (at least) were done in house... rather than through third parties. To ensure they met the advertising guidelines. Network sites, though... I think the ads are more third-party but still tightly restricted?
@Catija or drama...
@JoshZhang You don't need to reply, but have you tried to see if there is a pattern of some ads getting more "love"?
@Catija wasn't that changed some time ago and also meet with extreme backslash?
It depends and I'm still very new to this subject, I know some companies don't go through Google for various reasons but GAM is one of the main way we serve ads but I don't think it's the only way.
oh, don't worry, I was just thinking aloud.
Wondering if this is ad fraud or competitors stifling
3:33 PM
@SPArcheon-onstrike it's not that obvious sadly
My best guess is some bored troll looking for new ways to irritate SE and waste its developer's time.
@ShadowWizard seems stupid.
Maybe somebody they pissed off last Summer?
@SPArcheon-onstrike why? The goal is achieved. Those steps they take is the proof. :/
I mean, out of all the ways to hurt them this seems the least effective.
3:36 PM
That isn't cheap
@SPArcheon-onstrike not all trolls want to destroy SE, some just want to stir drama and havoc.
@JoshZhang hmm? What is that?
A click farm is a form of click fraud where a large group of low-paid workers are hired to click on links or buttons for the click fraudster (click farm master or click farmer). The workers click the links, surf the target website for a period of time, and possibly sign up for newsletters prior to clicking another link. For many of these workers, clicking on enough ads per day may increase their revenue substantially and may also be an alternative to other types of work. It is extremely difficult for an automated filter to detect this simulated traffic as fake because the visitor behavior appears...
brb going to make supper for kids
@ShadowWizard For some reasons they are investigating this, but it not like they HAD to. More clicks -> ads artificially looking more effective -> Company gets more money from (inflated) ads space selling. The only damage SE gets is increased bandwith usage and the like... but if that is your objective just play DDOS instead
here you are trolling and making them look like a better ads host at the same time
@SPArcheon-onstrike DDoS is boring at some point, and likely SE got really good fighting it.
3:39 PM
Why would someone angry at SE want to make them look like a better ads host?
this is probably unrelated to SE for once.
Except we don't look better. Companies are not in the business of paying for fake clicks.
@JoshZhang before that the company have to realize it, and I would assume anyone bad actor doing that wouldn't work on the assumption of getting caught.
Or are you implying they are doing this bad enough to get discovered on purpose and force you to waste resources on investigation?
@SPArcheon-onstrike I know that the ads that appeared more broadly network wide was problematic and led to issues. I guess what I don't know is whether the ads on tech sites are essentially a subset of SO ads or if they're third party but very, very tightly limited. I could imagine it's much easier to manage settings for a "tech" vertical of 3rd party as than trying to manage reasonable verticals for cooking, gaming, pets, etc.
I really can't tell what the motivation is, haven't really dug in that far. Maybe once I've got a better handle on everything I'll have time to comb through the logs and see if I can infer some kind of intent.
@JoshZhang oh, no pressure there, as I said just wondering because I found this somehow intriguing
3:43 PM
Also fake ad clicks is only one of the things we're dealing with. DDoS also never went away we get hit regularly. It's just we've had more time to get better at defending against those.
This interests me too, currently defending Stack feels like PvP
@JoshZhang To be fair, I have a fair bad sentiment about PVP. Having helped on an actual (albeit unofficial) mmorp server may have something to do with that.
Imho, most of the pvP is actually pvO
By what I mean Player VS Offline.
It lets me use a different skillset on top of what I need for the normal job so it's interesting. I played FPS and PvP'ed in MMO's for that reason.
@JoshZhang saw too much "wait at the dungeon gate to kill the players that just fought the boss inside and are low on potions and steal their loot"
Ah, WoW didn't have that specific situation but I can't say I wasn't a terrible teenager that did trolly / toxic things in video games.
I guess names like Barracoon the Piper aren't something you are familiar with
3:50 PM
UO was a little bit before I really got into online games
I started with OG Counter-Strike and an old Korean MMO
Trust me when I said that as little head the "WoW giant robot" had, UO had far less.
I've definitely heard stories from UO
@JoshZhang Can't remember the game for my sake, but considered the time period... let me guess, Wuxya style ?
Ragnarok Online
no wait, that is not the name
3:56 PM
Gotta grab lunch then some coding. Y'all can get Slate to summon me if needed.
As Josh has decreed, I shall now turn into a carrier pigeon.
oh, no it WAS Moonlight Blade.
I remember this trailer
basically "Wuxia infinite air jumping: the game"
Recently had a lot of games in BF2 ruined, because of two hackers just messing up the game. They weren't even playing, they just put everyone into a spectator-kind-of mode, and spammed the chat with "**** owns the BF servers". Some people just like to be idiots.
Back then all these Korean mmorpg had an huge merit.
05:00 - 16:0016:00 - 00:00

« first day (5161 days earlier)      last day (14 days later) »