ThioJoe

You might think differently if it was your face scammers were using to literally steal money from people directly below a video you worked hard on. The porno bots are one thing, they're a nuisance. The impersonators though, make my blood boil, and are the main reason I wrote the script in the first place.

Yea pretty much. It's in no way a be-all-end-all solution. There are definitely patterns spammers have been using for literally years that are so INCREDIBLY obvious that I can only conclude YouTube literally does not care at all. Those are the ones that the filter targets first.
 
Of course the spammers will try to get around it, and will soon enough. My thinking is not to stop them, but hopefully force them to use more and more convoluted means to get their message across, so that it makes their scams more obvious and less effective, therefore less profitable and less worth it.
 
Also I added multiple methods for searching for this exact reason. If you start noticing spammers using some technique to get around, you can just search for that themselves with one of the other modes.
 
 
I use the term 'obfuscated' really loosely. None of the code is obfuscated, only a few hardcoded keywords. I know the scammers that are at the top of the pyramid are going to figure it out anyway, so it's basically just to prevent the dumbest scammers (who just buy turn-key scam packs basically) from being able to quickly scan the code and look at the exact strings.

Yea pretty much. It's in no way a be-all-end-all solution. There are definitely patterns spammers have been using for literally years that are so INCREDIBLY obvious that I can only conclude YouTube literally does not care at all. Those are the ones that the filter targets first.
 
Of course the spammers will try to get around it, and will soon enough. My thinking is not to stop them, but hopefully force them to use more and more convoluted means to get their message across, so that it makes their scams more obvious and less effective, therefore less profitable and less worth it.
 
Also I added multiple methods for searching for this exact reason. If you start noticing spammers using some technique to get around, you can just search for that themselves with one of the other modes.
 
 
I use the term 'obfuscated' really loosely. None of the code is obfuscated, only a few hardcoded keywords. I know the scammers that are at the top of the pyramid are going to figure it out anyway, so it's basically just to prevent the dumbest scammers (who just buy turn-key scam packs basically) from being able to quickly scan the code and look at the exact strings.

You might think differently if it was your face scammers were using to literally steal money from people directly below a video you worked hard on. The porno bots are one thing, they're a nuisance. The impersonators though, make my blood boil, and are the main reason I wrote the script in the first place.

Yea pretty much. It's in no way a be-all-end-all solution. There are definitely patterns spammers have been using for literally years that are so INCREDIBLY obvious that I can only conclude YouTube literally does not care at all. Those are the ones that the filter targets first.
 
Of course the spammers will try to get around it, and will soon enough. My thinking is not to stop them, but hopefully force them to use more and more convoluted means to get their message across, so that it makes their scams more obvious and less effective, therefore less profitable and less worth it.
 
Also I added multiple methods for searching for this exact reason. If you start noticing spammers using some technique to get around, you can just search for that themselves with one of the other modes.
 
 
I use the term 'obfuscated' really loosely. None of the code is obfuscated, only a few hardcoded keywords. I know the scammers that are at the top of the pyramid are going to figure it out anyway, so it's basically just to prevent the dumbest scammers (who just buy turn-key scam packs basically) from being able to quickly scan the code and look at the exact strings.

Yea pretty much. It's in no way a be-all-end-all solution. There are definitely patterns spammers have been using for literally years that are so INCREDIBLY obvious that I can only conclude YouTube literally does not care at all. Those are the ones that the filter targets first.
 
Of course the spammers will try to get around it, and will soon enough. My thinking is not to stop them, but hopefully force them to use more and more convoluted means to get their message across, so that it makes their scams more obvious and less effective, therefore less profitable and less worth it.
 
Also I added multiple methods for searching for this exact reason. If you start noticing spammers using some technique to get around, you can just search for that themselves with one of the other modes.
 
 
I use the term 'obfuscated' really loosely. None of the code is obfuscated, only a few hardcoded keywords. I know the scammers that are at the top of the pyramid are going to figure it out anyway, so it's basically just to prevent the dumbest scammers (who just buy turn-key scam packs basically) from being able to quickly scan the code and look at the exact strings.

Dude what....? Do you even know what kind of bots everyone is referring to?
 
Anyone with half a brain cell can instantly spot them. Whether they're truly "bots" or someone copy-pasting the same message, is completely besides the point. 

Dude what....? Do you even know what kind of bots everyone is referring to?
 
Anyone with half a brain cell can instantly spot them. Whether they're truly "bots" or someone copy-pasting the same message, is completely besides the point. 

There's no doubt that's the ugliest function in the whole program and needs to be reworked, but in my defense it's also one of the first ones I wrote in the script, before I had even an inkling of an idea what I was doing (even though I still barely do). Until this program/script, the most 'advanced' thing I had ever written was this super basic script of ~100 lines (basically half of which are print statements) for replacing philips hue bulbs in that product's app. I've never said I was good at programming, and I am the first one to admit I'm crap, and I most definitely have never ever stated that the code would be elegant (just the opposite). But the first point I want to make here is that in the months since I wrote that function and others (which yes, is just one example of many hilariously ham-fisted coding 'strategies' in there) I've definitely picked up more knowledge and still continue to try to as much as I can, because I know my working knowledge is incredibly limited.
 
I am seriously just learning as I go, and the only reason I wrote this in the first place is because there were ZERO other options out there, and I wanted a tool for myself to use, and decided may as well put it out there for others. Believe me, I looked for any options. In fact at the beginning, it literally served one purpose, which was to just delete all comments by a single user on a video, because that's all I needed. For anything more complex, I wasn't getting nearly enough spam, for me it was always one or two bots who would come in and drop 50-80 comments that would take ages to delete, so I wrote the script just to delete them all at once. Then I realized that other channels were dealing with dozens of bots, where even that script was not very helpful, so I decided adding some ways to find certain spammers. Then I added a few more ways, and a few more still because I frankly found it fun to see it improve. What you see is not some grand-designed app intended to be a be-all-end-all spam killer. It's me repeatedly thinking, "oh but what if it also did this", googling only what's necessary to accomplish that, and me shoving that feature on top.
 
As for the "near 100%" accuracy claim, I should clarify. From my understanding, when testing for something there are two useful terms, "Sensitivity" and "Specificity". Which in this case basically distinguish between what percentage out of all the spammers it can detect (sensitivity), versus of those detected as positive, which ones are actually truly spammers and not false positives (specificity). I definitely jumped the gun in my video where I stated that it can detect nearly 100% of spammers, because I was only looking at whatsapp spammers (because that's all I ever saw on my channel). But have since learned about many more types which are harder to detect, and that I am indeed continuously working on. On the other hand, I am very confident in the program's "specificity" in the main 'auto smart' mode, which is basically the flagship filtering mode, and the only one I personally 'recommend' for general turn-key usage, because I have made the top priority for that mode to be effectively zero false positives (specificity), and sensitivity a close second. This is because if this thing is going to be used by potentially thousands of channels, the last thing I want is for it to be the cause of innocent commenters being caught on a large scale. The main mode is purely designed to be as general-purpose as possible with the understanding that screw-ups with it could cause major collateral damage, which  is why I still tell people to always confirm the comments before deleting. In fact, if I see even a single false positive I investigate and see how it can be eliminated. If the flagship mode does detect 100% of all scammers that's great, but at this point I obviously don't guarantee it, because that's not even necessarily it's purpose (which like I said is actually catching as many spammers with no false positives). I'd say it's more of a tool, not a solution.
 
And that being said, if a particular creator has a problem with other types of bots, there are four 'scanning modes' (what to scan) and with each you can use over a dozen 'filtering modes' (what to scan for) so that you can customize it completely to your needs. For example, there's filtering for individual characters / strings / regex, each within usernames / comment text / both, which totals 9. There's a more sensitive version of the auto smart mode which clearly states a much higher false positive rate. There's filtering for individual channel IDs (basically what the first version of the script did), and filtering usernames for a choice from 3 different non-ascii character sets.  The script can even scan and filter comments on community posts, which is not even supported by the API (making it ripe scamming grounds and spammers know it), so it has to resort to more 'brute force' methods I found. It's slower and clunky, but it works.
 
The whole app is not a perfect solution at all and I don't claim it to be, but I figure at least it's something, and as far as I can see it's all there is. I tried to add as many possible searching modes to cover every possible scenario, knowing that what I alone come up with isn't perfect. And yes, I know the program will certainly need to be continuously updated, and that dealing with scammers is a never ending cat & mouse game. 
 
Also, I'll point out that the spam list text files you see are actually only a recent and minor extra feature which just deals with the remaining edge cases of elusive spam comments that have few or zero notable signatures, besides the fact that they are commenting spam domains and such. It is not even close to being the main way it detects spam, it's just a brute force last-line of defense you might say. As for the blacklist type variables in the script itself, yes, those are the main bulk of the detection, but there is additional logic going, including the ability to convert unicode "confusable characters", therefore being able to see through any look-alike characters, of which spammers use all the time. 
 
To finish off this rambling I'll just say that really my main goal is not to eliminate scam comments completely. It's to at least make spammers have to rely on such convoluted and bizarre methods for getting around the filters, that they will become more and more obvious as spam comments, therefore hopefully clueing in most would-be victims.  The less it becomes worth their time, the less they'll do it, and that's literally the only way to stop them (in my opinion). At least on the channels that do use it, I think it would have a notable impact on whether spammers decide to hit that particular channel, and I have certainly seen a near complete disappearance of spam on my own channel.
 
All that being said, I'll reiterate that I don't pretend to be good at programming. I acknowledge the whole program is spaghetti to a hilarious degree. But I'm open to feedback and will continue to try and make the program as useful as possible within my abilities.
 
Thanks for listening to my TED talk.

Dude what....? Do you even know what kind of bots everyone is referring to?
 
Anyone with half a brain cell can instantly spot them. Whether they're truly "bots" or someone copy-pasting the same message, is completely besides the point. 

There's no doubt that's the ugliest function in the whole program and needs to be reworked, but in my defense it's also one of the first ones I wrote in the script, before I had even an inkling of an idea what I was doing (even though I still barely do). Until this program/script, the most 'advanced' thing I had ever written was this super basic script of ~100 lines (basically half of which are print statements) for replacing philips hue bulbs in that product's app. I've never said I was good at programming, and I am the first one to admit I'm crap, and I most definitely have never ever stated that the code would be elegant (just the opposite). But the first point I want to make here is that in the months since I wrote that function and others (which yes, is just one example of many hilariously ham-fisted coding 'strategies' in there) I've definitely picked up more knowledge and still continue to try to as much as I can, because I know my working knowledge is incredibly limited.
 
I am seriously just learning as I go, and the only reason I wrote this in the first place is because there were ZERO other options out there, and I wanted a tool for myself to use, and decided may as well put it out there for others. Believe me, I looked for any options. In fact at the beginning, it literally served one purpose, which was to just delete all comments by a single user on a video, because that's all I needed. For anything more complex, I wasn't getting nearly enough spam, for me it was always one or two bots who would come in and drop 50-80 comments that would take ages to delete, so I wrote the script just to delete them all at once. Then I realized that other channels were dealing with dozens of bots, where even that script was not very helpful, so I decided adding some ways to find certain spammers. Then I added a few more ways, and a few more still because I frankly found it fun to see it improve. What you see is not some grand-designed app intended to be a be-all-end-all spam killer. It's me repeatedly thinking, "oh but what if it also did this", googling only what's necessary to accomplish that, and me shoving that feature on top.
 
As for the "near 100%" accuracy claim, I should clarify. From my understanding, when testing for something there are two useful terms, "Sensitivity" and "Specificity". Which in this case basically distinguish between what percentage out of all the spammers it can detect (sensitivity), versus of those detected as positive, which ones are actually truly spammers and not false positives (specificity). I definitely jumped the gun in my video where I stated that it can detect nearly 100% of spammers, because I was only looking at whatsapp spammers (because that's all I ever saw on my channel). But have since learned about many more types which are harder to detect, and that I am indeed continuously working on. On the other hand, I am very confident in the program's "specificity" in the main 'auto smart' mode, which is basically the flagship filtering mode, and the only one I personally 'recommend' for general turn-key usage, because I have made the top priority for that mode to be effectively zero false positives (specificity), and sensitivity a close second. This is because if this thing is going to be used by potentially thousands of channels, the last thing I want is for it to be the cause of innocent commenters being caught on a large scale. The main mode is purely designed to be as general-purpose as possible with the understanding that screw-ups with it could cause major collateral damage, which  is why I still tell people to always confirm the comments before deleting. In fact, if I see even a single false positive I investigate and see how it can be eliminated. If the flagship mode does detect 100% of all scammers that's great, but at this point I obviously don't guarantee it, because that's not even necessarily it's purpose (which like I said is actually catching as many spammers with no false positives). I'd say it's more of a tool, not a solution.
 
And that being said, if a particular creator has a problem with other types of bots, there are four 'scanning modes' (what to scan) and with each you can use over a dozen 'filtering modes' (what to scan for) so that you can customize it completely to your needs. For example, there's filtering for individual characters / strings / regex, each within usernames / comment text / both, which totals 9. There's a more sensitive version of the auto smart mode which clearly states a much higher false positive rate. There's filtering for individual channel IDs (basically what the first version of the script did), and filtering usernames for a choice from 3 different non-ascii character sets.  The script can even scan and filter comments on community posts, which is not even supported by the API (making it ripe scamming grounds and spammers know it), so it has to resort to more 'brute force' methods I found. It's slower and clunky, but it works.
 
The whole app is not a perfect solution at all and I don't claim it to be, but I figure at least it's something, and as far as I can see it's all there is. I tried to add as many possible searching modes to cover every possible scenario, knowing that what I alone come up with isn't perfect. And yes, I know the program will certainly need to be continuously updated, and that dealing with scammers is a never ending cat & mouse game. 
 
Also, I'll point out that the spam list text files you see are actually only a recent and minor extra feature which just deals with the remaining edge cases of elusive spam comments that have few or zero notable signatures, besides the fact that they are commenting spam domains and such. It is not even close to being the main way it detects spam, it's just a brute force last-line of defense you might say. As for the blacklist type variables in the script itself, yes, those are the main bulk of the detection, but there is additional logic going, including the ability to convert unicode "confusable characters", therefore being able to see through any look-alike characters, of which spammers use all the time. 
 
To finish off this rambling I'll just say that really my main goal is not to eliminate scam comments completely. It's to at least make spammers have to rely on such convoluted and bizarre methods for getting around the filters, that they will become more and more obvious as spam comments, therefore hopefully clueing in most would-be victims.  The less it becomes worth their time, the less they'll do it, and that's literally the only way to stop them (in my opinion). At least on the channels that do use it, I think it would have a notable impact on whether spammers decide to hit that particular channel, and I have certainly seen a near complete disappearance of spam on my own channel.
 
All that being said, I'll reiterate that I don't pretend to be good at programming. I acknowledge the whole program is spaghetti to a hilarious degree. But I'm open to feedback and will continue to try and make the program as useful as possible within my abilities.
 
Thanks for listening to my TED talk.

There's no doubt that's the ugliest function in the whole program and needs to be reworked, but in my defense it's also one of the first ones I wrote in the script, before I had even an inkling of an idea what I was doing (even though I still barely do). Until this program/script, the most 'advanced' thing I had ever written was this super basic script of ~100 lines (basically half of which are print statements) for replacing philips hue bulbs in that product's app. I've never said I was good at programming, and I am the first one to admit I'm crap, and I most definitely have never ever stated that the code would be elegant (just the opposite). But the first point I want to make here is that in the months since I wrote that function and others (which yes, is just one example of many hilariously ham-fisted coding 'strategies' in there) I've definitely picked up more knowledge and still continue to try to as much as I can, because I know my working knowledge is incredibly limited.
 
I am seriously just learning as I go, and the only reason I wrote this in the first place is because there were ZERO other options out there, and I wanted a tool for myself to use, and decided may as well put it out there for others. Believe me, I looked for any options. In fact at the beginning, it literally served one purpose, which was to just delete all comments by a single user on a video, because that's all I needed. For anything more complex, I wasn't getting nearly enough spam, for me it was always one or two bots who would come in and drop 50-80 comments that would take ages to delete, so I wrote the script just to delete them all at once. Then I realized that other channels were dealing with dozens of bots, where even that script was not very helpful, so I decided adding some ways to find certain spammers. Then I added a few more ways, and a few more still because I frankly found it fun to see it improve. What you see is not some grand-designed app intended to be a be-all-end-all spam killer. It's me repeatedly thinking, "oh but what if it also did this", googling only what's necessary to accomplish that, and me shoving that feature on top.
 
As for the "near 100%" accuracy claim, I should clarify. From my understanding, when testing for something there are two useful terms, "Sensitivity" and "Specificity". Which in this case basically distinguish between what percentage out of all the spammers it can detect (sensitivity), versus of those detected as positive, which ones are actually truly spammers and not false positives (specificity). I definitely jumped the gun in my video where I stated that it can detect nearly 100% of spammers, because I was only looking at whatsapp spammers (because that's all I ever saw on my channel). But have since learned about many more types which are harder to detect, and that I am indeed continuously working on. On the other hand, I am very confident in the program's "specificity" in the main 'auto smart' mode, which is basically the flagship filtering mode, and the only one I personally 'recommend' for general turn-key usage, because I have made the top priority for that mode to be effectively zero false positives (specificity), and sensitivity a close second. This is because if this thing is going to be used by potentially thousands of channels, the last thing I want is for it to be the cause of innocent commenters being caught on a large scale. The main mode is purely designed to be as general-purpose as possible with the understanding that screw-ups with it could cause major collateral damage, which  is why I still tell people to always confirm the comments before deleting. In fact, if I see even a single false positive I investigate and see how it can be eliminated. If the flagship mode does detect 100% of all scammers that's great, but at this point I obviously don't guarantee it, because that's not even necessarily it's purpose (which like I said is actually catching as many spammers with no false positives). I'd say it's more of a tool, not a solution.
 
And that being said, if a particular creator has a problem with other types of bots, there are four 'scanning modes' (what to scan) and with each you can use over a dozen 'filtering modes' (what to scan for) so that you can customize it completely to your needs. For example, there's filtering for individual characters / strings / regex, each within usernames / comment text / both, which totals 9. There's a more sensitive version of the auto smart mode which clearly states a much higher false positive rate. There's filtering for individual channel IDs (basically what the first version of the script did), and filtering usernames for a choice from 3 different non-ascii character sets.  The script can even scan and filter comments on community posts, which is not even supported by the API (making it ripe scamming grounds and spammers know it), so it has to resort to more 'brute force' methods I found. It's slower and clunky, but it works.
 
The whole app is not a perfect solution at all and I don't claim it to be, but I figure at least it's something, and as far as I can see it's all there is. I tried to add as many possible searching modes to cover every possible scenario, knowing that what I alone come up with isn't perfect. And yes, I know the program will certainly need to be continuously updated, and that dealing with scammers is a never ending cat & mouse game. 
 
Also, I'll point out that the spam list text files you see are actually only a recent and minor extra feature which just deals with the remaining edge cases of elusive spam comments that have few or zero notable signatures, besides the fact that they are commenting spam domains and such. It is not even close to being the main way it detects spam, it's just a brute force last-line of defense you might say. As for the blacklist type variables in the script itself, yes, those are the main bulk of the detection, but there is additional logic going, including the ability to convert unicode "confusable characters", therefore being able to see through any look-alike characters, of which spammers use all the time. 
 
To finish off this rambling I'll just say that really my main goal is not to eliminate scam comments completely. It's to at least make spammers have to rely on such convoluted and bizarre methods for getting around the filters, that they will become more and more obvious as spam comments, therefore hopefully clueing in most would-be victims.  The less it becomes worth their time, the less they'll do it, and that's literally the only way to stop them (in my opinion). At least on the channels that do use it, I think it would have a notable impact on whether spammers decide to hit that particular channel, and I have certainly seen a near complete disappearance of spam on my own channel.
 
All that being said, I'll reiterate that I don't pretend to be good at programming. I acknowledge the whole program is spaghetti to a hilarious degree. But I'm open to feedback and will continue to try and make the program as useful as possible within my abilities.
 
Thanks for listening to my TED talk.

Dude what....? Do you even know what kind of bots everyone is referring to?
 
Anyone with half a brain cell can instantly spot them. Whether they're truly "bots" or someone copy-pasting the same message, is completely besides the point. 

Dude what....? Do you even know what kind of bots everyone is referring to?
 
Anyone with half a brain cell can instantly spot them. Whether they're truly "bots" or someone copy-pasting the same message, is completely besides the point. 

I watch quite a few tech/etc. youtubers, and am subscribed to several, including:
$300 Data Recovery Actually Hardcore Overclocking AMD (or do they count?) Austin Evans Barnacules Nerdgasm Bitwit (mostly on Floatplane) Carey Holzman der8auer Fractal Design (or do vendors / manufacturers count?) Gamers Nexus GNSteve Hardware Unboxed Hardware Canucks HDD Recovery Services iJustine (+ gaming, and just noticed I'm subbed to ijustinesiphone & otherijustine which haven't had uploads in several years) Jarrod'sTechss JayzTwoCents Joanne Tech Lover Krystal Key Level1Techs LGR Linus Tech Tips (mostly on Floatplane) LinusCatTips (Gotta watch his supervisors too!) Louis Rossman LowSpecGamer Luke Lafreniere Maraksot78 Mark Furneaux Moore's Law Is Dead NCIX Tech Tips Nerd on a Budget Nerdphilia Newegg Studios Nixie Pixel Noisy Butters NothingButTech Oldtech81 Paul's Hardware PC Perspective PCPartPicker PhilsComputerLab Professer Messer RandomGamingInHD RogueTech Gaming Sarah Dietschy Greg Salazar (I thought it was called "Science Studio"?) SouZy Taran Van Hemert TastyPC Tech Deals (mostly on Floatplane) Tech of Tomorrow Tech YES City TechLinked Techquickie The Tech Guy (I listen to his terrestrial radio show, mostly on Saturdays on KFI out of Los Angeles.  Church (or football) conflicts with Sundays so I usually can't listen then.) ThioJoe TWiT Tech Podcast Network UFD Tech The bold ones are ones I watch more often.  (Also a few of the other ones haven't uploaded videos in a year or more.  I left off a tech channel or a few, like MCS Tech, that I'm subbed to but took down their content.)
The ones I watch frequently vs occasionally seem to change at random.
 

Both would help, it's a matter of finding a sweet spot between the two. Both Level1Techs and ThioJoe have videos explaining the relationship, and providing spreadsheets to compute combinations by yourself - I don't know the links from the top of my head, though.
 
Own experience with FX chips seemed to indicate that 2400CL11 and 2133CL10 were very close in a certain benchmark, with 1866CL9 not far behind. But can't say whether that benchmark was representative of real world experience at all (my intuition tells me that real life experience won't care all that much in the big scheme of things, but FX chips need all the help you can give them - and free performance is free ).

@ThioJoe

I drink it all the time now, some days it's all I eat. I recommend ordering one case of the regular stuff, it's like $40, they also have the chocolate one which is good and tastes like YooHoo.
 
My Experience:
-First time I had one, it didn't taste bad, it just didn't taste like anything really.  I started drinking it and it was very difficult to drink the whole thing.  Body just wasn't used to eating/drinking something with such neutral flavor. Took me several hours to drink the first bottle, not because of the taste, because of the texture I guess. Was worried I wasted my money and wouldn't be able to drink the rest. My recommendation: Don't force it, don't drink more than you want. It will just be unpleasant and you'll get conditioned against it.  Also make sure it's really cold, probably makes it easier. Next day, I waited until I was really hungry and tried again.  This time it was a bit easier.  I was able to drink one bottle reluctantly in about an hour. Didn't have the same negative reaction like the first day. Day after that was even easier, I think my body realized "oh wait this shit is actually good for you". I might have even craved it a bit. After several days I guess (now a couple weeks later), it's not a problem to drink a whole bottle in a matter of maybe 10-15 minutes, faster if I really wanted I guess. I never got to the point of thinking it tastes "good", but I do actually crave it a bit.  But to be realistic, not as much as I'd crave a big juicy hamburger at the same time.  
Pros: Very convenient, easy to actually eat healthy. I hate cooking so much, and before I would usually wait until I was practically starving, then finally eat the fastest thing I could make, like a bowl of oatmeal or a microwave meal.  The chocolate one tastes really good but I still mostly have the regular kind because I wasn't sure if the chocolate one had anything extra in it (felt like there had to be a tradeoff for the better flavor, but I don't think there is).
 
Cons: Each bottle is only 400 calories, so you need to drink 5 a day to fulfill the 2000 calorie thing.  To be honest, it doesn't really taste good enough to want to drink that much. Usually I end up only drinking 4, then some other snack. It's kind of annoying being hungry just a couple hours after drinking one.  Also, it's not incredibly filling, but I guess technically that doesn't matter.
 
There are other options such as "Joylent" that I've tried. I actually think that tastes way better (there are several flavors), and you only need to drink 3 of those a day. It also has more protein and uses whey protein (before this edit I incorrectly thought it had primarily soy protein).  I'm pretty weary about soy in general, since the jury is still out on whether it is bad for men to consume a lot of it. Both soylent and joylent have some soy ingredients.
 

A lot of YouTubers had people bookmark their affiliate link, and I even looked into doing it myself. But reading through the Amazon Associate terms and conditions, it's very obvious that asking people to use an affiliate link as a bookmark is against the rules. Also, from what I understand, they keep track of the pages the links are linked from, and using a bookmark would not even give an affiliate commission since there is no referral page.

lol seriously?
 
I personally think it's both hilarious and sad ...
 
hilarious because I think they are funny videos, and I somehow find it funny that people really believe what youtubers like Thiojoe put on their channels!
 
sad because people really believe what youtubers like Thiojoe put on their channels ... and that it just proves that nobody is trying to learn anymore and only want to be spoonfed! (I personally can't stand people that are expecting to be spoonfed, that must be why I think it's hilarious they are getting "scammed" by those channels!)

You should be able to set your channel name as a "business name".  I believe your issue is that you may have your channel linked to a Google+ profile instead of a Google+ page.  I'm not sure how the current system works, because I know YouTube was removing the requirement of a Google+ link. HOWEVER, I know that previously, deleting a Google+ profile would result in deleting the channel as well, and they were supposed to change that, but I don't know if they did.  So definitely don't delete your Google+ profile.

The difference in input lag would probably be negligible compared to the inherently high input lag you get with consoles, namely 30fps, wireless controller, etc.

I've built several PCs but never a RAID server. In fact (quite embarassingly I've never used RAID before).  Basically I am not sure where to start, since the requirements for a server are so different from a PC.  This server would not be used as a final backup solution, but rather for video footage storage while working, and short term storage. Final backup would be onto LTO drives.
 
I know I want enterprise grade drives with UREs at least 10^15, but beyond that not much.  So what I don't know about is:
 
CPU: Probably a Xeon Motherboard: Never shopped for a server mobo before, not sure. Memory: I'm assuming something with ECC, but not sure how much RAID Controller: No clue about good brands / types here either OS: Not sure, maybe FreeNAS? My main consideration is driver compatibility for any interface cards I'd be installing now and in the future, such as Thunderbolt III or whatever. Array: Probably going to use RAID 6, but was thinking about ZFS because I hear good things about it.  
Anyone want to point me in the right direction?

Oh I didn't even notice the dude was being sarcastic lol. I was a bit surprised when he said cheap and budget friendly and linked to that camera. I was like, "maybe he just has a big budget."

That hurts my heart