Elon Musk modified the timeline to ensure his tweets came up first. Musk Response. UPDATE with Data to prove it.

[HenrySalayne]

1 hour ago, mr moose said:

And yes,  the evidence exists if it's easy enough to dig out musks twitter numbers then it can be done for any other twitter user.  No one is limited to seeing just musks data and only a week of it at that.

If it's so easy then go ahead, do that and post it here. You have done nothing so far but demanding that somebody else does the work. Why should they?

"I don't believe you because I don't have seen all this other data." is a sorry excuse. And even if somebody posts more data supporting this argument, we all know you will simply demand more and more things because "sample bias", "not comparable", "not applicable", "don't like the haircut", etc..

OP has posted a theory and then an source with data supporting this theory. If you found a source that disproves it, post it and be done with it. It's pretty simple.

1 hour ago, mr moose said:

Seeing as you and the OP are hell bent on this data being proof then you should have no problem pulling the rest of it and showing us that only musk benefited and is still benefiting.   Do that and I will concur with your hypothesis, failing to do that is just making shit up to explain an observation.

Please, feel free to quote the exact paragraph showing that I am "hell bent this data is proof". Where have I said that?

If you haven't noticed so far, I'm more of a meta-participant in this discussion. One party in this discussion has done nothing so far but scanning "we need evidence", "we need more evidence", "we need different evidence", "we don't believe this evidence" without contributing anything themselves. That's simply bad sportsmanship and adds nothing to this discussion.

[wanderingfool2]

On 2/28/2023 at 4:40 AM, HenrySalayne said:

By deduction this argument is proven.

theory --> observation

Musk asked for a a modification so his posts would benefit --> his posts have benefited

The ball is in your court now. Time to use some of this elbow grease yourself.

Then you would fail any first year philosophy class or first year CS predicate logic class.  You are trying to prove an antecedent with a consequent.  Logic does not work like that, to quote, "correlation does not imply causation".

 

There isn't any global warming -> record cold snap here

Therefore global warming isn't a thing. /s

 

Otherwise, I can equally claim that the Dec period where there was a spike was him getting internally boosted.  Or all the other accounts that get randomly more views for a few days are "boosted".

 

Like I said, there are other explanations that are grounded in things that occurred PRIOR to the news article.  Such as the algo using block list using count instead of by %, or the fanout service being bogged down when Musk tweeted (both which were mentioned before the superbowl as being an issue internally).

 

On 2/28/2023 at 4:40 AM, HenrySalayne said:

You must be delusional if you think asking for a master thesis grade paper on a forum is not moving the goal posts. Setting requirements so far out of reach just means one thing: you are not interested in a discussion, you just want to claim your superiority, because nobody could fulfil these requirements, thus they were wrong.

If you actually want to be right, find the data to proof it!

There isn't any proof given though.  You don't need a master thesis, one simply needs to show proper evidence.

 

The original article talks about having documents and sources...yet no direct quotes that relate to the claim and no documents showing to what is being claimed.  There's the activity of Musks' account, but if you are claiming it only affected Musks account you NEED to show other similar accounts to establish a baseline.  The simple fact as well, there are other scenarios that match what happened...ones that are backed by the tweets listed before this happened.

 

  

5 hours ago, HenrySalayne said:

If it's so easy then go ahead, do that and post it here. You have done nothing so far but demanding that somebody else does the work. Why should they?

Funny, I gave that kind of information earlier where I had stated the accounts.

[WolframaticAlpha]

On 2/19/2023 at 10:55 PM, Uttamattamakin said:

I'd quote more but paywall makes the source not very useful.  If you are interested in this story really and have the money take a look at it.  

On a tangent, this annoys the hell out of me. The news journals are the one place where they might benefit from microtransactions, but no, you need to buy a 10 dollar a month subscription to read two fucking articles. How annoying!

[HenrySalayne]

57 minutes ago, wanderingfool2 said:

Funny, I gave that kind of information earlier where I had stated the accounts.

Oh, did you? Because you mention it several times but you never actually put in a link.

"Trust me bro, I have also some really cool data showing exactly my opinion that I have posted somewhere, but nobody knows where not even me!"

 

 

Here is the complete encyclopaedia of your posts, maybe you can help me find it:

On 2/15/2023 at 4:59 PM, wanderingfool2 said:

So this is just another "rumor" of why with an "anonymous source".

 

Remember, it's also the press that brought you this wonderful gem
https://www.opindia.com/2022/10/prankster-conned-cnbc-ex-twitter-employee-fired-elon-musk-ligma-balls-meme-origin/

 

The article conveniently ignores the fact that people found that you get more impressions simply by switching your account to protected.  Specifically doing that doubled the impression count (as reported by other people).  The way I view it though, it's like when people discovered that subscribing to YouTube (pre-bell icon) that it wasn't recommending videos it thought you might not like (despite subscribing to the account).

 

My guess on what has happened, when the double the impression count was found out (Feb 1) there was a push to fix it.  Then it's realizing that it wasn't fixed after a few weeks asked why.

 

So even if one wants to believe the story of the "accounts" from employees; I think a more reasonable argument could be made that the following happened.  People notice you could get double the impressions by setting account to private, Musk tells engineers to fix it, engineers give excuse that his impression count is due to google (which looks like an excuse to justify the lower numbers without actually addressing the issue regarding the doubling), and now it's an attempt to see what impressions should be like.

 

Notice in the article that the claim is 90% of people who follow see the tweet...which really implies they filtered out tweets of people you follow (which for myself where I follow people who I want to see tweets of).

 

  

I'm pretty sure he is referring to it being a private company now, instead of a publicly traded company.  Publicly traded companies usually have to toe the line a lot more when it comes to certain aspects.

 

On 2/15/2023 at 6:09 PM, wanderingfool2 said:

Ignore the last bit why don't you, regarding that it could be changes to see what it should look like.

 

He literally uses his account to experiment and test with.  His account was set to private for a full day to test engagements (because again, people found it's double).  It's literally being able to test at scale (instead of simulating the load on the back-end).

 

Do you think it's not coincidental that things like this are happening right after someone figured out they got double the engagement from setting your feed to protected

 

On 2/15/2023 at 8:44 PM, wanderingfool2 said:

It's injecting pure speculation though based on "anonymous sources". 

 

Like honestly, around Feb 1 it came to light that there was a problem regarding showing current tweets to people (when you compare protected vs public accounts).  So it's a reasonable assumption that it would be getting looked into.

 

The news article even on the second paragraph talks about an internal memo which literally talks about debugging engagement.  So yea, less likely about the whining "I want more viewers" and more likely about "Why is engagement so messed up".  Again, you need to view it in light of the fact that this was started when a person tested "engagement" and found sending out a protected tweet got twice as many views (to which was replied roughly along the lines of that shouldn't be happening lets test and fix it).

 

So it is clear that they were looking at engagements and trying to figure things out (which yes, does likely mean messing around with some stuff live and potentially breaking a few things while it is getting fixed)

 

  

You can look for yourself.  It happened around Feb 1.  So it is highly important to note that there was an issue with engagements and it showing tweets to people who followed.  Don't believe me

https://arstechnica.com/gadgets/2023/02/musk-goes-private-on-twitter-to-test-if-locked-tweets-get-more-views/

 

 

On 2/15/2023 at 9:44 PM, wanderingfool2 said:

Define "reliable media sources", when it comes to things surrounding Musk and his companies though.  The fact is all that has been shown is that they are changing the way engagements are happening.  You can't conclude that "bigger than the President" type of claim.  Again before the news of all this broke, it was clear that they were starting to look into engagements.  Plus you injected politics by referencing not being able to run and how it's a good thing that he can't.

 

When most news sites decides to jump on one source and just parrot what they are saying, can one really call things reliable.

If you want a perfect example, look at the following stories from an April 17 crash (which NHTSA confirmed autopilot was NOT active, and after the accident Tesla stated it wasn't active...yet "anonymous sources", eye witness accounts confirmed it was active)

https://www.cnn.com/2021/04/28/cars/tesla-texas-crash-autopilot/index.html

https://www.washingtonpost.com/nation/2021/04/19/tesla-texas-driverless-crash/

https://www.cnbc.com/2021/04/18/no-one-was-driving-in-tesla-crash-that-killed-two-men-in-spring-texas-report.html

https://www.reuters.com/business/autos-transportation/us-probes-fatal-tesla-crash-believed-be-driverless-2021-04-19/

 

All that can really be said is that Twitter is attempting changes with engagements (and btw the most recent Barack Obama tweet had a surge in views compared to the last few weeks, given roughly the same amount of followers).  Unless some of the "anonymous" sources identify or provide actual emails (instead of hearsay) then it's just a rumor.

 

Your conclusion and the articles conclusion is that he's doing it essentially because he's just petty, when there is actual evidence showing that they were working on engagements prior to the so call "incident" that lead to the recent change.

 

On 2/16/2023 at 7:49 AM, wanderingfool2 said:

News isn't meant to be read and treated as though what they are saying is fact.  A lot of "news" is simply created where "facts" are omitted or presented in deceiving ways.  The other times you posted and I responded to is exactly like that (presenting facts that don't tell the whole story).

 

Again, I posted 4 articles that all spat out the same thing (and never bothered to update their article when it was clear that lies were told).

 

The simple fact is that there was a problem with engagements, which again was clear from the Feb 1 stuff.  It's an ASSUMPTION, not a fact that the recent spike was caused because of the superbowl tweet and an ASSUMPTION that it's an ego thing...instead of actual concern that the engagement is broken.

 

If I follow someone I want to see what they are tweeting, not what the algorithm decides which posts I want to see...if someone tweets out something and only a small fraction sees the tweets it's a problem.  That's the whole reason the Feb 1 thing got started, and likely has lead to the experimentation.  If an account with 6x the amount of followers gets less impressions on a tweet when tweeting about the same thing then yes, that does show that there is a big issue with how the engagement is being done; and does warrant a look into.

 

Again, it's making an ASSUPTION, while ignoring everything that seem to counter it.  So let's point out again things that counter it

1) KNOWN FACT: Privating your account increased impressions (Feb 1) [in some cases 5x] and notifications to followers started working.  With Musk privating his account for a day to verify the results.  This alone really shows that they were looking into it before

2) MUSK TWEET: Feb 11 (so prior to the SuperBowl), statement that the fanout service was bogged down when Musk tweeted resulting in 90% of tweets not even being displayed (so likely issue with scale).  Talks about how block count (blocking accounts) are used when determining recommended instead of block %

3) MUSK TWEET: Feb 12 PRE SuperBowl...talked about tracking "dropped tweets" (so likely something from fanout service again)

4) Based on Musk's tweets, he deleted his entire blocked list (The tweet came from Feb 13)...see point 2

 

So yea, all the SuperBowl thing likely did was show that there was still a problem.

 

  

See point 2 and point 4.  A tweet about block count being used (instead of block % based on followers), and then sometimes recently him deleting his entire block list.  The likely hood is that he is testing on his account (and maybe the block-list count was used in more than a few places)

 

On 2/16/2023 at 2:50 PM, wanderingfool2 said:

The first line of the first article I wrote literally stated it was Tesla's statement (Tesla made no such statement). 

 

So YES I am comparing it because the articles were attributing statements not from random bystanders, but attributed towards Tesla.  Even the police went as far to make a statement that they thought no one was driving.  While I can't find them now (there are many that were updated later), but some of them also quoted "Tesla Insiders" as confirming it was on during the crash.

 

You are assuming it's broadcast to everyone, and assuming it's a visibility boost; vs making a single change to an account and seeing the effects (like lets say removing his list of blocked users)

 

"Researched journalism", where they offer no proof except that it's "inside sources" and offer a slack message that does nothing to prove they are boosting visibility.  If they have inside sources that know what happened, where are the slack messages that show it, where is the proof that visibility is boosted (because if it's boosted then someone should be able to have a screenshot or something)

 

The articles I posted earlier show more than clearly that click-bait and fearmongering work better than actually well researched journalism.

 

It's also not pure guesswork, it's an educated guess based on what was happening to the lead-up.  Like I said in my later post, we know they were looking into (Feb 1 was the first test of privating account), then Feb 11 talked about block list count instead of block list % was used to determine essentially the penalty for visibility, and sometime between Feb 12 - Feb 13 (when all this occurred) we know Musk said he deleted all his block list.

 

So yes, testing things involving visibility on a single large account is sensical in the sense that you can test certain scenarios...such as if they did push it to everyone, it could be to try stress-testing the fanout service which apparently had issues in the past (btw, notice how the article buried the fact they were talking about followers receiving his tweets).

 

On 2/20/2023 at 5:59 PM, wanderingfool2 said:

The only evidence they give is the "view count" (and back on the "anonymous sources" while also referencing the Plaftormer article).

 

Even if you assume it was true, you would expect the data to show the tweets from the 13 and beyond getting more tweets.  Yet it started on the 12th, with tweets prior to SuperBowl.  It was already admitted a bug was introduced to worked on the replies.

 

They still don't offer evidence to what the claim is that he said to modify it so he was given more weight.  It's a simple prepositional statement that they are trying, while failing to grasp a lack of understanding: if p then q, where they are claiming p = forced engagement, q is impression count.  Even if you prove q, you do not prove p, at best you have a correlation.  It doesn't also negate the ACTUAL evidence that exists that they were working on engagements prior to this. (Including the fact that the fanout server was being bogged down)

 

They also mention that engagements for Dec were in the 230 range average, yet their fancy bar graph shows the very tail end of Jan which shows half the engagements as Dec.  (so those spikes make it look worse than it is).

 

So offer evidence, not some "anonymous sources" which without direct quotes can twist what is being said.

 

Like for example, they talk about "bots"  in that article and how to eliminate them they hurt Ukraine (when they blocked carriers)...what they fail to mention is that they blocked carriers which had SMS bot spam contributing over 10% of bots (at least in a statement made).

 

Also, it's not reliable.  Your so called WA PO article is written by the same journalist from the one I posted earlier who couldn't even bother updating their article that it wasn't a driverless article.  The same journalist who claim in the headlines "4 hours to put out" (Fire chief stated it's a mischaracterization, as it was actually put out quickly...they couldn't move the vehicle due to the deaths so they couldn't get to the areas which cause occasional re-ignitions), and it wasn't a driverless vehicle (the driver exited after the crash).  So yea, I wouldn't put much credit on WA PO when it comes to vetting sources.

 

On 2/22/2023 at 7:43 PM, wanderingfool2 said:

The only "proof" so far is an increase in impressions over superbowl weekend (along the same time pieces came out about Musk, which would have also driven impressions).  The simple fact is, the "journalist" is citing "people familiar with the situation" (Not even trying to claim people who were there), not giving any direct quotes (except for things that are irrelevant to the claim), and claims he has supporting documents (which isn't shown).

 

So yea, no proof that what is being claimed as the headline was even true.  Your other WA PO source, references the platformer article, as I pointed out the author already has slanted previous articles without updating them when 100% irrefutable facts are presented.  Your Yahoo source is based on the Platformer one.  So yea, it comes down to do you trust a journalist who makes a living literally by making click-bait titles that he vetted his sources.

 

As an example, if you look at period Dec 16 - Dec 18 you get about 1.4 billion impressions vs Feb 12 - Feb 14 1.34 billion.  [It's what starts happening when you start cherry picking data that "meets" the requirements for their narrative). 

Compare that to Dec 19 - Dec 21, you get 470 million.

  

Last I checked, we are still allowed calling your sources garbage in regards to them actually showing anything.  It's our opinions that your source shows no proof of what is claimed.

 

On 2/26/2023 at 7:31 AM, wanderingfool2 said:

It doesn't prove the anonymous source.  First, it brings me back to the stats I listed before.

 

If you look at period Dec 16 - Dec 18 you get about 1.4 billion impressions vs Feb 12 - Feb 14 1.34 billion

Compare that to Dec 19 - Dec 21, you get 470 million

 

I could claim over the 3 day stretches that Dec 16 - 18 there was a 300% drop in impressions compared to Dec 19 - 21.  Actually, if I wanted to get really nasty in terms of "stats", Dec 18 vs 19 saw over a 1500% drop.

 

OR you could look at another account that has similar follower count.   The day after superbowl vs previous tweet...oh look an account with 133 mill followers...tweet  one day after superbowl = 2.9 mill impressions, tweet the very next day = 17.8 mill.   So 613% increase.

 

Also, as I mentioned prior to all of this there was already talk about the fan service being bogged down, and that blocking users caused the algorithm to essentially tank your numbers. (Followed by talks around the time of SuperBowl that Musk deleted his block list)

 

p.s. Mashable is also the mind numbing publisher who also decided to write (and make a big deal) about how Tesla wasn't subscribed to Twitter Blue...despite Tesla being subscribe to Twitter Gold (the one for businesses...but convenient that they leave details like that out)

 

On 2/27/2023 at 7:28 AM, wanderingfool2 said:

Well the terminology I was using was wrong, but the general concept that I was using to compare like numbers was still sound.  When someone is using an increase of 700% a relative drop where the reference point is the high number would be 87.5% decrease which muddies things up when people look at the numbers.  So instead I based it on the lower number, but didn't want to get accused of clickbaity kind of thing.  The general point still got across in what I wrote.

 

Evidence does not equal sources.  I've already mentioned the other account before with similar size, and mentioned how there was a surge of impressions at the same time.  So yea, I'm not going to quote the same types of sources

 

One can simply also scroll back to look at twitter numbers, and can see that it's quite clearly that a short term analytics increase doesn't mean a thing, given you had a similar thing happen in December.  (Seeing surges in impressions doesn't equate the anonymous source being true, especially when there were tweets prior to that that would also justify the surge, including but not limited to the news linking to Musk's twitter).

 

NO IT DOESN'T

 

The data you showed can be related to an enormous number of reasons, INCLUDING the changes to his block list which were being stated.  Also including the glitch that they already recognized that was promoting replies as well.

 

It's like climate deniers saying that since we are having a record cold Jan/Feb this year global warming mustn't be a thing.  Like I said as well, you have the Dec period which saw massive amounts of engagements on a few days.  (Followed by lack luster the very next 3 day period).

 

The simple fact is, there are plenty of reasons (and some documented before this article) why engagements would be changing/debugging it...so no seeing a change in impressions doesn't prove your article correct in the anonymous statement.

 

The simple fact is, the author of it claims he has documents (provided none), and hides behind people who are in on the situation (while not giving any direct quotes).  Or you know what, Barack Obama must have also demanded being buffed in the algorithm because his impressions increased by 6.13x. from the 13th to 14th.

 

On 2/27/2023 at 7:18 PM, wanderingfool2 said:

It's not moving the goal posts, it's literally pointed out that the evidence being shown is faulty.  Like even if you want to claim that it was an abnormal spike, it's basic logic to show that the spike didn't also occur on other similar sized accounts.  There is also the general lack of sound logic being applied in the sense that showing a spike in impressions does not prove that a superbowl tweet and Musk getting mad is true.

 

The simple fact is, the counter argument is that the source/evidence is garbage.  If I claim the global isn't warming because we had record snowfall, a valid counter pointis stating that local snowfall isn't indicative of the global average trend.

 

On 2/27/2023 at 8:18 PM, wanderingfool2 said:

Then don't get mad when people call the story that was published garbage tabloid.  If you have the power to prove it, then prove it.  Or is this another whining about how Norton was compromised because people used credential stuffing.  Just because you want to believe it's true doesn't mean you showed it was true.

 

Basic statistics is if you are saying something is not natural, and it's only affecting one account you show examples of it not affecting other similar accounts.

 

It's not that hard of a concept, otherwise again, IT'S SNOWING HERE GLOBAL WARMING DOESN'T EXIST.  Showing a singular event that could have multitudes of causes, and in isolation of showing the same data over the same period of time on similar samples is just bad statistics.  Again global warming example, every day I could point to a city that breaks a 25 year old record for being cold (what is missing is that there are 10,000 cities so statistically there will likely be 1 city that has a record cord)

 

57 minutes ago, wanderingfool2 said:

Then you would fail any first year philosophy class or first year CS predicate logic class.  You are trying to prove an antecedent with a consequent.  Logic does not work like that, to quote, "correlation does not imply causation".

 

There isn't any global warming -> record cold snap here

Therefore global warming isn't a thing. /s

 

Otherwise, I can equally claim that the Dec period where there was a spike was him getting internally boosted.  Or all the other accounts that get randomly more views for a few days are "boosted".

 

Like I said, there are other explanations that are grounded in things that occurred PRIOR to the news article.  Such as the algo using block list using count instead of by %, or the fanout service being bogged down when Musk tweeted (both which were mentioned before the superbowl as being an issue internally).

 

There isn't any proof given though.  You don't need a master thesis, one simply needs to show proper evidence.

 

The original article talks about having documents and sources...yet no direct quotes that relate to the claim and no documents showing to what is being claimed.  There's the activity of Musks' account, but if you are claiming it only affected Musks account you NEED to show other similar accounts to establish a baseline.  The simple fact as well, there are other scenarios that match what happened...ones that are backed by the tweets listed before this happened.

 

  

Funny, I gave that kind of information earlier where I had stated the accounts.

 

[wanderingfool2]

6 minutes ago, HenrySalayne said:

Oh, did you? Because you mention it several times but you never actually put in a link.

"Trust me bro, I have also some really cool data showing exactly my opinion that I have posted somewhere, but nobody knows where not even me!"

Again, if you bothered to read you would realize the data was pulled from the twitter feed with the analytics to show in comparison that the claimed impression count for the superbowl wasn't exactly something new having a spike in views.  Also the other accounted compared against again was Obama's account which has a similar follow account...and again, you see a spike from one day to the next...so according to your ignorant logic then Oboma must have gotten preferential treatment.

 

So yea, that is a source.  It's something that anyone can verify, unlike the evidence "here's the hourly of Musk's and the trust me bro all other accounts didn't exhibit the same thing".

[Bitter]

Good gravy,it's like Twitter in this thread!

[HenrySalayne]

15 minutes ago, wanderingfool2 said:

Again, if you bothered to read you would realize the data was pulled from the twitter feed with the analytics to show in comparison that the claimed impression count for the superbowl wasn't exactly something new having a spike in views.  Also the other accounted compared against again was Obama's account which has a similar follow account...and again, you see a spike from one day to the next...

Again, If you already said it, why do you bother re-writing it instead of pointing where I can find it?

 

16 minutes ago, wanderingfool2 said:

so according to your ignorant logic then Oboma must have gotten preferential treatment.

If you bother to read, you would realize it is not according to my logic. ¯\_(ツ)_/¯

20 minutes ago, wanderingfool2 said:

So yea, that is a source.

No, it's not. We are literally still at "trust me bro", there is no reference in your post linking to any source.

[Uttamattamakin]

2 hours ago, wanderingfool2 said:

Again, if you bothered to read you would realize the data was pulled from the twitter feed with the analytics to show in comparison that the claimed impression count for the superbowl wasn't exactly something new having a spike in views.  Also the other accounted compared against again was Obama's account which has a similar follow 

 

 

The SPike is not the issue.  The issue is the Horizontal line in the data where he was getting artificial engagement over time.  A spike is a natural thing to see.  Everyone gets a spike when they tweet.  

 

If it was a spike you'd be right, but it's not a spike... it's something that would only happen if external force was applied in a slapdash way.  Then the fact the spikes afterwards are higher is due to Elons tweets after the fact being enhanced compared to the past.  Only his got this.    

Also why pick Obama?  Is it ok to mention presidents here now?  Obama gets a lot of engagement the same reason that other politicians do.  Presidents and heads of state matter in a way that even the richest man on Earth never ever will. 

[wanderingfool2]

22 minutes ago, Uttamattamakin said:

The SPike is not the issue.  The issue is the Horizontal line in the data where he was getting artificial engagement over time.  A spike is a natural thing to see.  Everyone gets a spike when they tweet.  

A horizontal line in their "data" shows absolutely nothing in regards to "what" is happening.

 

Since you seem so hung up on terrible data, look at the chart and all those "nice" straight lines that there are at approximate the same intervals.  You want to know why it's a straight line, because their data collection points stopped and they literally are playing connect the dots.  You can tell that because from 10th to 11th they had a similar 18 hour (approx) stretch where it's a solid straight line (sloped down, but it's a solid line) vs ones where they do have data points 11th to 12th (during a 6 hour period approx you can see the ups/downs).

 

So yes, you are right, the plateau isn't natural, but that's because they are lacking datapoints.  So their data is garbage at that, since they claim they sampled at 1 hour intervals, yet it's clear that they aren't sampling at 1 hour intervals otherwise they wouldn't have that many straight lines.  From what I can tell they had 6 data points from 11 - 12.  8 from 12-13, 9 from 13-14...but none of the datapoints are fixed in terms of how spread out they are.  So that in itself screams that the data isn't valid on such a short timescale.

 

35 minutes ago, Uttamattamakin said:

Also why pick Obama?  Is it ok to mention presidents here now?  Obama gets a lot of engagement the same reason that other politicians do.  Presidents and heads of state matter in a way that even the richest man on Earth never ever will. 

I've already stated, approximately the same follower account at 133 mill followers.

[Uttamattamakin]

1 hour ago, wanderingfool2 said:

A horizontal line in their "data" shows absolutely nothing in regards to "what" is happening.

 

Since you seem so hung up on terrible data, look at the chart and all those "nice" straight lines that there are at approximate the same intervals.  You want to know why it's a straight line, b

Not just straight.  Horizontal. Indicating the exact same number of interactions per time interval with NO fluctuation, no noise.  Thats not natural and yes, I am an expert on this sort of thing. But whatever man no reasonable data I could post here  and not violate the guidelines would satisfy you. 

[wanderingfool2]

7 hours ago, Uttamattamakin said:

Not just straight.  Horizontal. Indicating the exact same number of interactions per time interval with NO fluctuation, no noise.  Thats not natural and yes, I am an expert on this sort of thing. But whatever man no reasonable data I could post here  and not violate the guidelines would satisfy you. 

If you are an "expert" in this sort of thing then you should be fired immediately.  It's horizontal because they only had like 2 data points that were pretty similar in size.  You can tell that the datapoints are contrived by how the 10th to 11th had a straight line (albeit downwards sloped).  That means the data they collected is not "hourly" as they claim, or at the very least you can't get the granularity at 1 hour intervals.

 

Here's a hint, "1 hour intervals" WILL NOT SHOW UP AS STRAIGHT UP LINES FOR MULTIPLE HOURS ON END.  No matter if you are pushing it to more people, you will see fluctuations as more or less people are using the platform.  Even if the article wasn't a bunch of unsubstantiated rumors, it would NOT cause a straight horizontal line, if it truly was hourly you would see wobbles in the data; the fact you don't clearly indicates that it does.

 

So again, it's back to what I said in my previous post about that data, it's garbage data because they claim it's hourly but clearly it's not actual hourly impressions (or rather the data isn't granular enough to base things off of hourly).

 

If you have things that prove it then post it, don't hide behind some made up story that you can't post data that proves it.

[mr moose]

@HenrySalayne  Look up "burden of proof"

 

 

[Isuck Assimov]

Hhehehe, TSLA goes

[Uttamattamakin]

3 hours ago, mr moose said:

@HenrySalayne  Look up "burden of proof"

 

 

The burden of proof is not static.  One side presents proof then to disprove it you need more or better data.  

[Uttamattamakin]

On 3/2/2023 at 8:37 AM, wanderingfool2 said:

If you are an "expert" in this sort of thing then you should be fired immediately.  It's horizontal because they only had like 2 data points that were pretty similar in size. 

If what you said was true then there would be a lot of flat horizontal lines on that chart.  There are not.  

To get that even on an hourly basis would require 2 hours of Elon's tweets getting EXACTLY the same (not just pretty similar) engagement.  That's even more damming 

On 3/2/2023 at 8:37 AM, wanderingfool2 said:

You can tell that the datapoints are contrived by how the 10th to 11th had a straight line (albeit downwards sloped).  That means the data they collected is not "hourly" as they claim, or at the very least you can't get the granularity at 1 hour intervals.

You ....Horizontal not just straight.  A line going up or down on a chart like this is a natural fluctuation with time as a random distribution of data, a natural distribution of data would predict.  A horizontal (or even perfectly vertical line like a stair step) is not a natural occurrence.  THen you can see how all the peaks are higher after it.  All of that due to Elon artificially boosting his tweets. 

 

< removed by moderation >

As to the rest of this

 

On 3/2/2023 at 8:37 AM, wanderingfool2 said:

Here's a hint, "1 hour intervals" WILL NOT SHOW UP AS STRAIGHT UP LINES FOR MULTIPLE HOURS ON END.  No matter if you are pushing it to more people, you will see fluctuations as more or less people are using the platform.  Even if the article wasn't a bunch of unsubstantiated rumors, it would NOT cause a straight horizontal line, if it truly was hourly you would see wobbles in the data; the fact you don't clearly indicates that it does.

 

So again, it's back to what I said in my previous post about that data, it's garbage data because they claim it's hourly but clearly it's not actual hourly impressions (or rather the data isn't granular enough to base things off of hourly).

 

If you have things that prove it then post it, don't hide behind some made up story that you can't post data that proves it.

Edited March 4, 2023 by LogicalDrm

[wanderingfool2]

2 hours ago, Uttamattamakin said:

You ....Horizontal not just straight.  A line going up or down on a chart like this is a natural fluctuation with time as a random distribution of data, a natural distribution of data would predict.  A horizontal (or even perfectly vertical line like a stair step) is not a natural occurrence.  THen you can see how all the peaks are higher after it.  All of that due to Elon artificially boosting his tweets. 

Look at the data point which I called straight (between 10th and 11th), it's not horizontal the bit I was pointing out it was straight (sloped downwards).  Since you can't seem to piece two and to together, if you grab 2 data points and draw a line between them you form a straight line with a slope.  If you grab 2 data points and they happen to have similar y axis levels then it's also a straight horizontal line.

 

A line over a period of 8 HOURS, no matter what should have natural fluctuations up and down just due to Twitter usage.  Usage by hour fluctuates, so even he was boosted, it wouldn't produce a horizontal line.  Instead it's back to what I said, that 8 hour period is just easily explained that there was only 2 data points and the person drawing the graph just connects the 2 data points.

 

2 hours ago, Uttamattamakin said:

If what you said was true then there would be a lot of flat horizontal lines on that chart.  There are not.  

To get that even on an hourly basis would require 2 hours of Elon's tweets getting EXACTLY the same (not just pretty similar) engagement.  That's even more damming 

I already showed you an area on the graph where it only has 2 data points over a similar time period.  It's not hard to realize they are missing data points, remember they claim it's hourly sampling.  Since you can't seem to understand that I already gave you a period in your "proof" which shows something is wrong with the way the data was collected.  Let me show you visually

On the "hourly" graph, there is at least 2 time periods where they had 2 points of data over "8 hours", and about 4 time periods that had 2 points over "4 hours". 

So yea, garbage data and garbage time-frame of events that they claim.  They claim an area is 8 hours long, based on the bottom section there 24 hour period is at least 32 hours....but that doesn't fit their narrative.  You can see that it would have to over over 12 hours (closer to maybe 13 hours).  Feel free to try it yourself, duplicate that period that they claim is 2:30 - 10:30 and watch how there is no way it could be just 8 hours (and has to be over a 12 hour period). 

 

2 hours ago, Uttamattamakin said:

That data fits the story and back up the anonymous source. A journalistic practice that is time honored and respected by journalist.

Except it doesn't, see above it's garbage data and they couldn't even bother to make it align to the alleged story.  Again, look at prior days analytics (not over a course of a month), you find stretches where it exhibited massive spikes in the past before.  You find also that you have massive drops.  You find that a similar sized account on Feb 14 had a similar spike compared to Feb 13 (in terms of ratio).  The fact is you and the journalist haven't offered any proof to corroborate the claim, it's literally just saying "trust me I'm a journalist and have the data but won't show it".

[Bitter]

If we can't trust Musk, can we trust ANY data from Twitter?

[WhitetailAni]

On 3/1/2023 at 6:24 AM, HenrySalayne said:

If it's so easy then go ahead, do that and post it here. You have done nothing so far but demanding that somebody else does the work. Why should they?

If you make a claim, it's on you to prove it. You do not get to force your opponent to do it for you.

 

They have set their metric for what counts as proof for them. If you don't want to meet it that's fine, but that comes with the downside of them not believing you. Choose one.

[HenrySalayne]

8 hours ago, mr moose said:

@HenrySalayne  Look up "burden of proof"

Ok, we can end the debate at this point. Your statement proves conclusively that you are unfit to distinguish between an online forum and a court room. I rest my case.

 

[WhitetailAni]

Just now, HenrySalayne said:

Ok, we can end the debate at this point. Your statement proves conclusively that you are unfit to distinguish between an online forum and a court room. I rest my case.

 

Why would the rules of argument change. I'm not proving your argument if I'm arguing against you, that goes against my whole goal.

[HenrySalayne]

29 minutes ago, FakeKGB said:

If you make a claim, it's on you to prove it. You do not get to force your opponent to do it for you.

 

They have set their metric for what counts as proof for them. If you don't want to meet it that's fine, but that comes with the downside of them not believing you. Choose one.

26 minutes ago, FakeKGB said:

Why would the rules of argument change. I'm not proving your argument if I'm arguing against you, that goes against my whole goal.

I think you missed a few bits in your assessment.

OP made a claim and proved this claim by deduction (not legally proven but enough for this debate to move forward).

The other parties involved are claiming

- that they don't find the proof sufficient (which is an utterly pointless claim that can be made over and over and over and over again ) or

- that they have some opposing sources somewhere in their previous posts, where I couldn't find anything nor could they themselves find anything and point at it.

One party tried at least to proof their claim that the evidence provided by OP is not sufficient (and failed in supplying a source or pointing at it if there was a source) while the other party couldn't even do that. Instead they spun up this laughable requirement catalogue (for an online debate

Quote

In order to prove something was changed to favor Musk you have to show that there was nothing else that caused or effected the change and that only musk was effected positively, otherwise all you have is a thing that changed and either everyone was effected positivity or there were compounding events that caused it.    You know how burden of proof works right?  You claim it was done with the intention to inflate musk's presence on twitter, you now have to prove that all other factors had no influence on that change (which includes proving no one else ecperienced the same change to their impressions/likes/retweets/whatever metric you feel is most important).

tl;dr

'Only if everything, even the hand of god, has been shown to not have interfered, you have proven to the world that Elon Musk ordered a modification of Twitter in his favour."

Spoiler

tl;dr

'You cannot proof that. so my opinion is better than yours.'

 

Nobody was asked to defend the others parties' claim or change their believes, they were just asked to act a little less like stubborn children. Children, who probably think that providing a source would open them up for other people challenging their claims...

 

 

[wanderingfool2]

7 minutes ago, HenrySalayne said:

OP made a claim and proved this claim by deduction (not legally proven but enough for this debate to move forward).

It is not deduction.  It's a logical fallacy that you are trying to apply.  If p then q, even if you try claiming he showed q that doesn't prove p.

 

So no he proved nothing, and like I showed above as well the only "evidence" that was presented is massively flawed and outright wrong (claiming an 12+ hour period is somehow only 8 hours).  So yea, you are wrong in that it proves anything.

 

 

7 minutes ago, HenrySalayne said:

One party tried at least to proof their claim that the evidence provided by OP is not sufficient (and failed in supplying a source or pointing at it if there was a source)

ARE YOU TRYING TO IGNORE what I said.  LOOK at the data you literally find on data on the accounts mentioned on the dates mention.  But yea, you keep saying that I didn't provide any sources yet you can clearly go look at those dates if you want to verify that the numbers I provided are roughly correct.

[Bitter]

I feel we're getting closer and closer to Godwin's Law and kind of surprised we haven't hit it yet.

[HenrySalayne]

16 minutes ago, wanderingfool2 said:

LOOK at the data you literally find on data on the accounts mentioned on the dates mention.

But where?

 

 

Spoiler

+1 for slapstick comedy, made me burst out laughing.

 

 

13 minutes ago, Bitter said:

I feel we're getting closer and closer to Godwin's Law and kind of surprised we haven't hit it yet.

You mean 'incompetence and arrogance go hand in hand'?

[wanderingfool2]

1 minute ago, HenrySalayne said:

But where?

 

 

  Reveal hidden contents

+1 for slapstick comedy, made me burst out laughing.

 

 You literally don't know where to find the data...when it's a topic about Twitter, and I mention looking at impressions.  How about going to twitter, and the account and the date and lets see...looking at the impressions.

On 2/26/2023 at 10:28 PM, wanderingfool2 said:

Or you know what, Barack Obama must have also demanded being buffed in the algorithm because his impressions increased by 6.13x. from the 13th to 14th.

Since you can't be bothered here's the Feb 13 and here is the Feb 14.  Under your logic of "deduction" I can claim that Obama must have been buffed; since if I make the claim he did there is my "proof".  Which goes to show your powers of deduction.  The reason you look at an account like Obama as well is because it's literally 133 mill vs 130 mill followers.

 

You are more than welcome to go to every Musk tweet (not reply) to verify that Dec range as well...which I already said.  Just because you are too lazy to look at the source doesn't mean I didn't provide sources.