The premise doesn't hold up when you think about it. If to be great is to be good repeatably, then what does it mean to be great repeatably? Is that also great or do we need a new word?
Producing good work consistently means you are good overall. Being great means you produce great work consistently.
One of the linked articles in the piece contradicts the notion that consistently being good would lead to greatness. (Well a link to a tweet of images of the article -- https://autotranslucence.com/2018/03/30/becoming-a-magician/). That author argued that while they were consistently very good at body painting (top 5 in the world championships), they were not able to conceive how the 2016 winner produced the work they did in the same amount of time (6 hours). Even though they were improving incrementally, they were so far behind the winner's work that they couldn't wrap their head around what sorts of improvements they would need to add in order to reach the same level of competency. So it was not just a matter of being consistent. Their mental model was off.
That seems to fit reality and greatness more than the original article. A good mathematician does not simply reach the level of Terence Tao by being consistently good over time.
The word "Great" has to be qualified here. Reading the article, I don't believe the author means to say you can "work" your way to actual greatness - say being the best in the world at something, or one of the world's leading experts in a tough field.
Rather greatness here is simply being close to one's true potential. Whether you are Lebron James or a random bench-warmer, the trick is to show up everyday and consistently improve. LeBron is great because his floor is higher than most other player's ceilings. There are many players who can give a brilliant performance on their day, but can't do much on others. If you've watched any sport, the great players are very very good even on their worst days. On their best days, they are unbeatable.
If you can identify the area where you can achieve the same and put in the work, then you're on the path to greatness (at your level).
The article is arguing just that. There is no great consistently and yes you do achieve whatever level Terence Tao is at by being good and improving consistently.
Software is a particularly good form of this. You can't really just vomit out a codebase that's perfect immediately. If we could do that then there wouldn't be so many versions of even the most mundane tools on your computer.
Even though when you look at a great piece of spftware you can just look ay the whole completed thing, it didn't just pop into existence that way. It was honed over time.
> That author argued that while they were consistently very good at body painting (top 5 in the world championships), they were not able to conceive how the 2016 winner produced the work they did in the same amount of time (6 hours). Even though they were improving incrementally, they were so far behind the winner's work that they couldn't wrap their head around what sorts of improvements they would need to add in order to reach the same level of competency. So it was not just a matter of being consistent. Their mental model was off.
I think that that author would be able to improve her article by many orders of magnitudes simply by incorporating the images she is referring to (his own, and the winner's).
That she is incapable of understanding how poor a written comparison of pictures is when making a point of comparison tells me that she might have difficulty understanding things in general, not just the visual art she is, ironically, only able to talk about and not show.
Isn’t cache invalidation the hardest part of the feed? Once a user with millions of followers tweets, doesn’t that essentially invalidate millions of caches? That doesn’t sound like a trivial problem.
Because you can't edit tweets, the database is monotonically increasing (append-only) so scaling it isn't that hard. Cache invalidation basically just adds a new item. Deleting a tweet actually would be the harder/more expensive operation, but it's also less common.
Editing a tweet is also easy because the feeds would only store its id. You'd only need to invalidate one cache where the tweet object itself is stored.
WhatsApp and twitter’s latency calculations are on different things.
Twitter’s latency stems from calculating what tweets should show on a given request. Even if you try to show tweets from 1 minute ago, it’s hard to cache that stuff using traditional systems because of the fan out. If an account with 50 million followers tweets, you need to update 50 million timelines. How do you do that quickly?
And you would have to define maximum latency, is it seconds, minutes, hours? because you can’t have the timelines be inconsistent for too long as that leads to some people getting news faster than others.
now you have to deliver them, exactly one time, to each recipient or groups of recipients, through different network topologies, with different challenges and vastly different bandwidth and latency guarantees, in exact order, while also keeping track of who is online e who's not, and distributing that information in real time, only to the edge nodes that should know about it, all of that fully E2E encrypted but stored (indefinitely?) in case the recipient is currently offline and unless that recipient blocked the sender.
let's agree that both companies solve hard problems and that it's not the technical difficulties that make the two companies sizes so different.
> This stuff is staggeringly, mind-wreckingly expensive for people who aren't in tech, and yet functionally required because of the structures we have allowed to be built.
Society has always worked this way. Those who have rare skills get paid the most. Supply and demand and what not. Universities are gateways to advanced skills, especially in traditional occupations where equipment is often expensive (medical, chemical, mechanical, etc). The reason you go to a university is so that you can get advanced skills in order to make an advanced salary. It makes no sense to go to a university by default and come out with a degree that doesn't teach you advanced skills that get you a high wage. If the jobs that your degree are going to get you aren't going to pay for what that degree cost you then you made a poor decision by taking on that debt.
This sort of thing is why I believe basic economics should be a hard requirement in high school. You shouldn't be able to get a high school diploma without understanding the mechanisms of debt/leverage. So many people have screwed themselves over because they don't understand that the only reason to ever take on debt is to use it as leverage so that you can earn even more than the debt you took on. Any other reason is foolish.
It's really sad when you think about it, so many people would be way better off if they knew the definition of leverage. Such a simple concept, yet so powerful (it's funny how knowing about leverage gives one so much leverage in life).
> The reason you go to a university is so that you can get advanced skills in order to make an advanced salary.
Most of your post is pretty good, but I laughed aloud at this, tbh. The reason you go to a university is because your resume gets thrown out for almost any desk job--hell, for Starbucks--if you don't have a bachelor's.
It is functionally necessary. These aren't "rare skills". These are employer-mandated minimums, and it leaves people with that inflated student debt, encouraged and pushed upon them by their parents and by the expectations of society, to subsidize those employers' demands.
> I'm graduating with my master's in CS now and still haven't been taught in class many of the things I had to figure out to make what I wanted to make.
Maybe the fact that you figured it out is an indication that it doesn't have to be taught in a CS curriculum. Most of the things that you mentioned have great documentation and tutorials outside of the classroom. The fact that so many people use them without them being taught in schools should be proof enough. I think the current situation makes sense -- Computer Science education should focus on teaching the fundamentals of algorithms, data structures, discrete math and how to think conceptually about problems without regard to implementation. By learning about data structures and the like, you indirectly learn how best to use memory, redis, MySQL, write servers and whatever else.
I often receive flak for expressing a similar opinion when people express disappointment at CS not teaching them "how to code". Your tuition is best spent on conceptually difficult material, not stuff you can easily pick up on your own.
Whenever I talk to people about data science I repeat the old adage:
A data scientist is a statistician who can program or a programmer who understands statistics at a high level.
It's amazing how few CS people become data scientists, I think physics and economics degrees are the most common degrees among elite DSs I know.
but it actually makes a lot of sense - it's relatively easy to pick up coding. In fact, it could be the best-taught skill on the internet, there's no shortage of material for a person who wants to learn to code. I think that most CS people don't learn enough statistics to really pursue DS and a ton of people with advanced stats knowledge can pickup coding relatively easily.
Now when I talk to kids, I tell them that if they want to go to college, they should try to learn something they can't learn on their own.
I'm surprised the author's takeaway from the Joyce quote was disillusionment. He's actually quite close to the secret sauce. I don't think writing more code and ending up with less code are at odds. In fact, I think you need to write more code to end up with less. You write more code in order to understand the problem space. Once you understand the problem space, you can then refactor everything that you wrote into more concise code that more accurately reflects the problem.
James Joyce wrote Ulysses at the rate of a hundred words per day if you only consider the finished product. However, I doubt every word that Joyce wrote ended up in Ulysses. I'm sure he did quite a lot of cutting and rewriting.
The takeaway from me from the two quoted stories was that quality and quantity are not at odds, but instead are the yin and yang of productivity. They reinforce each other. The more things you produce, the more patterns you are exposed to which in turn leads to higher quality since experience leads to efficiency, giving you more time to get things right.
I think I get it, but I'd suggest sticking with the same example when explaining the confrontational aspect of it. When you switch contexts it makes it hard to imagine the inverse. If you don't mind, can you explain the confrontation aspect in the context of getting onto the rollercoaster?
> Depriving people of their livelihood for expressing political views is absolutely a free speech issue.
I don't think free speech is the issue here. There's a difference between expressing your views on your own time and involving the company that you work for to adopt those views. We've all seen the disclaimers that people put on their twitter profiles and blogs distancing their personal views from the companies they work for. Such a disclaimer was not issued here. Colin involved the 49ers Franchise and the NFL in his political views. He used their broadcasts to send out his message. Not all the owners necessarily sympathized with his views, especially when it started bringing negative attention to their business. In that regard, I don't think it's all that surprising that it cost him his job.
Note also that there are plenty of current NFL players who express the same sentiments as Colin on social media and in interviews, and they still have their jobs.
Yes free speech means the government can't limit what you say (with some specific exceptions). It doesn't mean you won't face consequences or backlash or insults or derision from those in the civilian world who disagree with you.
Audio books help a lot. It makes reading a more passive activity that you can do while doing other things that are less taxing on the mind. If you have a long commute and you can knock off quite a few books just going back and forth to work. My commute is around 30 minutes, which means I can listen to at least 5 hours of audio each week. I also tend to listen to audio books while doing chores -- tending the yard, cleaning the garage, cleaning the house, buying groceries, etc.
Most nonfiction books tend to be around 8-9 hours so you can usually finish one book a week, maybe 2 or 3 if you listen at 1.5x. If it's scientifically backed with a lot of studies and arguments, then it's closer to 15-18 hours. Novels are all over the place, from a short 4 hours to epic 50 hours books. But since it's all passive, it becomes less about finishing because it's easy reading. You'll finish it eventually.
Audio books are also great in the sense that they save my eyes for things that I can't listen to passively, like technical or mathematical books.
