10x engineers most certainly exist, I've had the fortune of being able to work with a few and their true impact isn't in the code they themselves write but in the higher-level strategic/architectural decisions they make that impact the team as a whole. That being said it is also my experience that most self-described 10x engineers are anything but, so if one bases their impression of 10x engineers off those self-promoting cases then I can see how it would appear to be a myth
I mean in a sense traditional record labels are already exactly that for the music industry, it's the same financing and payoff model (record labels front an advance to the artist and recoup the advance and more on album sales/revenue share from touring, 90% of signed artists will likely flop but the 10% that take off could become the next Beyonce or Justin Bieber)
Congrats on the launch, a question on the technical side of things -- how much computational overhead does your desktop client incur on the host machine? If you're running CV algorithms locally, would that necessitate higher minimum system performance specs beyond those recommended by Riot/other game vendors to avoid impacting the performance of the game client itself?
I have yet to deep-dive into the techical details of Pegasus so this is pure speculation, but if Pegasus demonstrates the feasibility of deploying and executing arbitrary payloads on remote devices without the owner's knowledge or consent wouldn't that raise a blackmail vector with this new program? Again I'm not familiar with the specific exploits used by Pegasus, but is there anything stopping another malicious actor from relying on the same exploits to say install a bootloader that downloads images that would trigger a positive scan?
Given the tone of the rest of the letter I don't think that statement can be taken at face value, it comes across much more like a backhanded jibe adding insult to injury
I think a lot of the muffled aesthetic is a consequence of the decisions made in the recording/mixing process, there's a strong room reverb across the recording particularly noticeable on Ringo's drums which combined with the high-end softening of the Fairchild 660 compressor featured on Beatles records gives it a "rock concert while standing in line for the bathroom" color. I could be mistaken but it sounds like a lot of the instrumental parts were recorded in single takes in the same room compared to say Abbey Road where it sounds like they were tracked independently. A consequence of tracking multiple lines together is there can be cross-instrument fuzzing that results in "muffling", it can be an intentional effect though for a more live sound. Just my 2c
Going with your proposal for a decentralized Airbnb/Uber -- how would you handle dispute resolution and trust mediation without a centralized authority in an IRL setting, off-chain? If you have a bad actor in the system, how would you screen/remove them and conversely, how would you prevent abuse of the screening mechanism you propose? Relying on the blockchain as a decentralized database/audit record is one thing, but I think IRL enforcement and mediation of off-chain matters are critical prerequisite matters that need to be addressed before these types of marketplace businesses can even be considered feasible as dApps
I have a few counterpoints to those made in the article:
- As much as current developments have made for more accommodating work/social environments for introverts, I wouldn't go as far as to say that "introverts have the upper hand". If anything, I would say the shift towards increasingly digital modes of interaction tends to reward those with a flair for self-advocacy even more -- just think about who tends to dominate discourse in online forums/social media/conference calls. I personally rely heavily on non-verbal cues to express myself/gauge others in social settings but there isn't really a channel for that type of expression in digital environments. That in turn leads to me feeling like I need to be more verbally engaged despite that not being my natural inclination, which then feeds back into a spiral of mental exhaustion -> degraded conversation and thinking -> more exhaustion.
- The characterization of extroverts as shmoozers and social climbers and introverts as doing "actual work" is rather unfair -- in the past, I've worked with both "introverts" who recognized they could skate by just doing the bare minimum to say they did something in standup and staying out of the way of everyone else and "extroverts" who would actively drop by to check if I needed help with anything. Moreover, in roles like that of a technical lead or solutions architect, being able to clearly articulate and discuss designs is a far more important skill than being a code-cranking steam engine which by the definitions of the article are "extroverted" qualities. When it comes to implementation I'd agree that heads-down, interruption-free environments are the most conducive, but the further I progress as a software engineer the more I've come to realize that implementation is only the tip of the iceberg.
I think introverts as characterized in the article do have it as good as ever as employees, but that is still far from a "reign of introverts". With the growing trends of creator economies and self-employment, self-advocacy is only going to become an even more crucial skill set and mentality to develop -- I can anecdotally attest that the notion that dedication to your craft alone is enough to carry you is a fallacy, and I think those most prone to burnout/depression in creative and self-employed endeavors are those who don't anticipate the level of self-advocacy needed to break through.
Had a relatively similar experience (finally understood what drinking out of a firehose meant), the silver lining of the experience though is it finally dispelled any notion of imposter syndrome when I realized everyone was running around as much of a headless chicken as myself ahah
Same with the UC system, if you look at the Asian/Asian-American demographic it's in some cases close to double that of comparable non-race blind institutions. When I was going through the college application process in my day, the unspoken Asian/Asian-American quotas for top private institutions were accepted as a given rule of the game which led to some pretty vicious rivalries among my peer group (whether accurate or not, there was a notion that our district was allocated a sub-quota of the overall quota, so it was perceived to be a zero-sum game as to who would get HYPSM offers each year). Even if the recent court rulings weren't in their favor, it's heartening to see newer cohorts of Asian-American applicants start to challenge the rules directly instead of solely focusing on their internal competition
Talk about Baader-Meinhof, was just wondering last night how my messages get routed to friends and relatives overseas. Fascinating find, thanks for sharing
Man I should've done my research before joining, my team was regularly pulling 70+ hour weeks including weekends for the first four/five months I was there (granted we were understaffed and behind schedule for launch at re:Invent at the time). Grateful for the experience because everything else feels like a breeze afterwards, but also definitely never putting up with that again lol
I find alternative data to be a fascinating world, your idea reminds me of CargoMetrics which is doing a similar thing in the shipping industry. Just wondering -- if and when you reach the scale where a substantial segment of the supply side of the electricity market is using your service, is there the possibility that your algorithms can be subject to a reverse self-fulfilling prophecy (e.g. you predict a downwards demand shock which leads producers relying on your forecasts to scale back generation, but because so many producers are doing so in tandem there ends up being a power shortage)? Also, are you exclusively targeting the supply side, or do you have plans for a consumer-side offering as well? I feel like there would be immense potential for an integrated "smart" power grid especially with the rise of planned smart cities in the Asia-Pacific region so that might be another interesting application of your forecasting. Congrats on the launch, looking forward to hearing where y'all go from here.
Former AWS dev here -- the bar is the same but the workloads can be so insane that you meet the requirements faster than anywhere else in the company (there was one engineer that made principal like 5-6 years out of undergrad, which means they would've had to have been promoted every 1-2 years assuming they were hired as an L4). I think parent comment might be referring to more lenient stack ranking policies, I never experienced any of it during my time there but then again I saw teams with 20-30% annual turnover rates so no need to force people out when they're already showing themselves the door I guess :)
I recall there was another study that observed sleep deprivation being associated with alleviation of depression symptoms, so I wonder if the same psychobiological mechanism is at play here (if waking up early could be scientifically considered as a mild form of sleep deprivation). I also notice most of my peers working remotely and myself included tend to wake up later in the day and I wonder how much if any of the widespread miasma during lockdowns/mandated WFH period could be attributed to this phenomenon -- there are plenty of confounding factors but I think it would be worth investigating further.
This is just a conjecture, but I think the 280 character limit on Twitter (or whatever it is now) is one of the key culprits behind the trend of deteriorating discourse not just at the societal level but even at a personal level. It's simply impossible to express nuance in a 280-character span and if you can't win over an audience within that limit, they'll just scroll past whatever else you have to say. Even if you personally have a more intricate perspective on the matter, you're forced to play by majority rules if you want to reach an audience and that quickly becomes a race to the bottom in a system that rewards whoever can come up with the best one-line mic drop instead of the best formulated ideas. I've noticed in my very limited experience with Twitter that it seems to drive "engagement" from opposing sides more so than Facebook's echo chamber, which is a breeding ground for confrontation (whereas long-form Facebook rants seem more so geared towards getting people on "your side" to egg you on). In theory our society needs that sort of cross-aisle engagement to reverse the trend of political fragmentation, but if all you have is 280 characters to make your case that just simply doesn't work. I wonder if there is a solution that could make it work, but given the trend of social media culture being driven by the lowest common denominator I'm inclined to think you'd need some sort of centralized moderating authority and that just brings us back full circle to traditional media outlets
Haven't read that book so I'm not familiar with the contents, but if you're interested in creating VST plugins you might want to check out some frameworks out there like JUCE: https://juce.com/
I don't know if you're a musician yourself but I'd highly recommend downloading a DAW and playing around with the plugins to get a sense of common audio processing modules, you can perhaps make it a near-term goal to create an EQ or filter plugin that you can install and use in the DAW. IIRC I think Ableton Lite should have a free demo, Reaper is another well-regarded free DAW from what I've heard
- You would need some sort of barrier preventing movement of researchers between these audit teams and the institutions they are supposed to audit otherwise there would be a perverse incentive for a researcher to provide favorable treatment to certain institutions in exchange for a guaranteed position at said institutions later on. You could have an internal audit team audit the audit team, but you quickly run into an infinitely recursive structure and we'd have to question whether there would even be sufficient resources to support anything more than the initial team to begin with.
- From my admittedly limited experience as an economics research assistant in undergrad, I understood replication studies to be considered low-value projects that are barely worth listing on a CV for a tenure-track academic. That in conjunction with the aforementioned movement barrier would make such an auditing researcher position a career dead-end, which would then raise the question of which researchers would be willing to take on this role (though to be fair there would still be someone given the insane ratio of candidates in academia to available positions). The uncomfortable truth is that most researchers would likely jump at other opportunities if they are able to and this position would be a last resort for those who aren't able to land a gig elsewhere. I wouldn't doubt the ability of this pool of candidates to still perform quality work, but if some of them have an axe to grind (e.g. denied tenure, criticized in a peer review) that is another source of bias to be wary of as they are effectively being granted the leverage to cut off the lifeline for their rivals.
- You could implement a sort of academic jury duty to randomly select the members of this team to address the issues in the last point, which might be an interesting structure to consider further. I could still see conflict-of-interest issues being present especially if the panel members are actively involved in the field of research (and from what I've seen of academia, it's a bunch of high-intellect individuals playing by high school social rules lol) but it would at least address the incentive issue of self-selection. Perhaps some sort of election structure like this (https://en.wikipedia.org/wiki/Doge_of_Venice#:~:text=Thirty%....) could be used to filter out conflict of interest, but it would make selecting the panel a much more involved and time-consuming process.
Given the stories I've heard about the rigor of UChicago's econometrics program, I'm somewhat surprised they would put out a working paper extrapolating what is effectively a single data point to make broader inferences about the efficacy of WFH practices. To be fair they qualify their findings in terms of the firm's employees, but given the tone of the title and conclusion focusing on WFH as a general trend it still seems to me that they are drawing from a sample size of n=1 instead of n=10000. The paper doesn't specify which Asian country is being studied but anecdotally I have several relatives/friends working in East Asian corporations who can attest to the pervasiveness of "managerial piety" culture even when working remotely, so I would also think they'd want to replicate the study across similar companies in other parts of the world to account for potentially confounding cultural factors in assessing WFH productivity
Something I wonder about is whether the inflow of money into all these "assets" is indeed driven by people with buckets of excess capital, or actually from people who could use the cash in hand the most. Anecdotally I know individuals crushed by student loans/HCOL who immediately threw their stimulus checks into crypto and meme stocks with the rationale that if they lose it all, a few thousand dollars would've made no difference on their debt anyways whereas getting on the next rocketship trend early could become a life-changing sum of money. I agree something is fundamentally broken if we've reached the point where we live in a casino economy, I don't know how long it can last and the long-term potential consequences are quite concerning
I find it humbling to hear about how much goes on behind the scenes in different industries that we so often take for granted, I'd bet the average person probably hasn't thought much about the shipping industry until the whole Suez situation over the past week (including myself coming from a family of three generations of sailors haha). Correct me if I'm wrong but given the real-time nature of your data I'm assuming your software is meant to be accessed on board vessels in transit, so are outages/connectivity issues at sea an issue you've encountered? If so, do you have systems/protocols for handling prolonged outages (e.g. due to inclement weather or network interference) and reconciling your operational algorithms with potentially stale data?