This is such an obvious and important issue that it unironically influences what technology I use since these online noise generators (youtubers/bloggers) make it much harder to find valuable content.
In my personal experience it became much harder in the past few years to find useful in-depth information about rust/vim/linux (my "default" stack of choice) since these became more mainstream (you'd think that's good news..). The rate at which useful information is generated is much lower than the rate at which noise is created, and the noise is more popular for obvious reasons, making finding good information increasingly hard with time (since more views equals more visibility). Most of the good content I find (on youtube, which admittedly isn't the best place) always has less than 5000 views, and very often less than 500.
I've been thinking about this ever since you released that video explaining the exploit of the Apple's M-series chips. A lot of supposed experts on YouTube were making 10-15 minutes videos and not saying much more than "it's a problem in the hardware" while sometimes trying give brief explanations on some parts of modern CPUs. Then I watched your hour video where every 10 minutes you were saying "I am not an expert" while explaining in detail how the CPU parts interact with one another and end up manifesting the exploit. The difference between you and them is night and day! I know that those types of videos (like yours) probably don't get much views compared to the mainstream pop-science-y ones. This is the only explanation that I can give myself on why people are not investing more time and effort into making detailed in-depth explanations of these subjects -- "I can make, but no one would watch it".
All of this is to say -- thank you! You seem to have much harder job that pays back much less and I am glad that there are people like you that can pass on the in-depth knowledge they've acquired through laborious research and years of experience.
Another reason I can think of is that most viewers are not tech-interested, but they still want to get involved at some shallow level. So in-depth tech report is just too hard, and too long to chew. It also requires effort and commitment in the long run.
Only a minority number of viewers are truly going for the root cause.
Again, thanks Casey for bringing this gold mine of content here.
I'm a design verification engineer at Tenstorrent and a few months ago I was tasked with developing some tests for our side channel attack mitigation features. I didn't know anything about how attacks like Spectre and Meltdown worked so I looked around for some explainer videos. I found multiple videos that seemed to imply that the sensitive data was just left in the cache and the attacker process just spied on it. Even knowing nothing about it I was like "that... doesn't seem right." I had to go read the research papers to actually make sure I had a correct understanding.
I have become incredibly skeptical of most things. It is not unnreasonable to anticipate this in areas that you are unfamiliar with, when you see this stuff going on in an area you are familiar with.
It's easy to get sucked into the entertaining content of tech influencers. When a person listens and learns from them on topics they are not knowledgeable about, it's easy to consume incorrect information, similar to trusting ChatGPT when it generates responses that only sound reasonable to the uninformed. For example, instead of following YouTube tech influencers who merely repeat what they hear from others, professionals should look for YouTube channels that host professional and academic talks from true experts - not people trying to push a new video every other day. This post reminds us all of the importance of true knowledge and experience, avoiding the hype, resisting the urge to be first, and instead focusing on true understanding and clear communication so others can benefit from what we have to contribute.
Thank you for standing up for integrity. It's a win-win for everyone. I do believe you can make integrity and quality popular though. On top of your suggestions to proper creator/influencer strategies, it might also be important to briefly clarify these requirements to make an accurate statement, and why we have to be hesitant making statements without those facts, since I think this is a topic that reaches out much further than those knowledgeable of tech stack. I imagine there's many non-technical people who just want to know how it truly happened simply for the sake of knowing how to prevent or avoid exposing themselves to similar risks in the future. Or people who just need a break from their own concerns and need to point and laugh at drama details of sectors they don't know very well. Etc.
This is such an obvious and important issue that it unironically influences what technology I use since these online noise generators (youtubers/bloggers) make it much harder to find valuable content.
In my personal experience it became much harder in the past few years to find useful in-depth information about rust/vim/linux (my "default" stack of choice) since these became more mainstream (you'd think that's good news..). The rate at which useful information is generated is much lower than the rate at which noise is created, and the noise is more popular for obvious reasons, making finding good information increasingly hard with time (since more views equals more visibility). Most of the good content I find (on youtube, which admittedly isn't the best place) always has less than 5000 views, and very often less than 500.
I've been thinking about this ever since you released that video explaining the exploit of the Apple's M-series chips. A lot of supposed experts on YouTube were making 10-15 minutes videos and not saying much more than "it's a problem in the hardware" while sometimes trying give brief explanations on some parts of modern CPUs. Then I watched your hour video where every 10 minutes you were saying "I am not an expert" while explaining in detail how the CPU parts interact with one another and end up manifesting the exploit. The difference between you and them is night and day! I know that those types of videos (like yours) probably don't get much views compared to the mainstream pop-science-y ones. This is the only explanation that I can give myself on why people are not investing more time and effort into making detailed in-depth explanations of these subjects -- "I can make, but no one would watch it".
All of this is to say -- thank you! You seem to have much harder job that pays back much less and I am glad that there are people like you that can pass on the in-depth knowledge they've acquired through laborious research and years of experience.
Another reason I can think of is that most viewers are not tech-interested, but they still want to get involved at some shallow level. So in-depth tech report is just too hard, and too long to chew. It also requires effort and commitment in the long run.
Only a minority number of viewers are truly going for the root cause.
Again, thanks Casey for bringing this gold mine of content here.
I'm a design verification engineer at Tenstorrent and a few months ago I was tasked with developing some tests for our side channel attack mitigation features. I didn't know anything about how attacks like Spectre and Meltdown worked so I looked around for some explainer videos. I found multiple videos that seemed to imply that the sensitive data was just left in the cache and the attacker process just spied on it. Even knowing nothing about it I was like "that... doesn't seem right." I had to go read the research papers to actually make sure I had a correct understanding.
I have become incredibly skeptical of most things. It is not unnreasonable to anticipate this in areas that you are unfamiliar with, when you see this stuff going on in an area you are familiar with.
There is a quote attributed to Michael Crichton about this sort of thing, actually! It's about the opposite behavior, where people see things they know are false in their area of expertise, but then fail to realize that's happening outside their area as well: https://www.goodreads.com/quotes/65213-briefly-stated-the-gell-mann-amnesia-effect-is-as-follows-you
- Casey
It's easy to get sucked into the entertaining content of tech influencers. When a person listens and learns from them on topics they are not knowledgeable about, it's easy to consume incorrect information, similar to trusting ChatGPT when it generates responses that only sound reasonable to the uninformed. For example, instead of following YouTube tech influencers who merely repeat what they hear from others, professionals should look for YouTube channels that host professional and academic talks from true experts - not people trying to push a new video every other day. This post reminds us all of the importance of true knowledge and experience, avoiding the hype, resisting the urge to be first, and instead focusing on true understanding and clear communication so others can benefit from what we have to contribute.
Honest and in-depth content is like gold to me nowadays.
The job market is forcing everyone to claim to be an expert and stand out from the crowd. The boom of social media makes this easier than ever.
Such bad news for the tech environment. Hope this will somehow resolve soon.
Thank you for standing up for integrity. It's a win-win for everyone. I do believe you can make integrity and quality popular though. On top of your suggestions to proper creator/influencer strategies, it might also be important to briefly clarify these requirements to make an accurate statement, and why we have to be hesitant making statements without those facts, since I think this is a topic that reaches out much further than those knowledgeable of tech stack. I imagine there's many non-technical people who just want to know how it truly happened simply for the sake of knowing how to prevent or avoid exposing themselves to similar risks in the future. Or people who just need a break from their own concerns and need to point and laugh at drama details of sectors they don't know very well. Etc.