ChefboyOG

This. It's a cycle of bundled complexity to unbundled "simplicity" and back. I'm not so sure it's even entirely unique to software.

Predictive text and image classification are not the extent of ML's limits, even going by what is currently in production. Recommendation engines, ETA prediction, translation, drug discovery, medicinal chemistry, fraud detection—these are all areas where ML is already very important and present.

Sure, it's not artificial general intelligence, but what technological invention in history would compare to the impact of AGI? That's sort of a weird bar.

In theory, I think this is a plausible positive scenario. In practice, however, I've never seen it play out like this (huge anecdata here).

I've worked remote most of my career, and in my experience, the politics and work theaterisms are still there, just different.

For example, someone's perceived level of "effort" often becomes a function of how responsive they are. Do they respond to Slack messages at random hours, when team members in different time zones/aggressive managers ping them? Do they frequently communicate with the team via video calls and Slack discussions? Just like in normal office politics, their actual productivity or work quality is assessed less than their appearance of engagement.

That's just an example, but my experience in general has been that remote doesn't change anything about the cultural of management at a company. Whatever metrics mattered in person, there will be remote equivalents to replace them.

The positive here is that good leadership is still good leadership, remote or not. The negative, obviously, is that remote isn't the balm for poor leadership that many hope it will be.

I mean, corporations do regularly offshore as much work as they possibly can and build new plants/offices in low cost locations. Obviously, they need employees, but it's not as if companies don't already do everything they can to reduce the cost of labor.

I believe they're referring to buying lab equipment on eBay.

https://www.nature.com/articles/d41586-018-05655-3

I always wonder about bias in these statistics around age.

It makes intuitive sense to me that a cancer diagnosed in a young patient, who is below the common screening age, is probably being diagnosed because it is presenting serious symptoms (i.e. is growing fast).

Cancers diagnosed in a 70 year old, on the other hand, would seem much more likely to be diagnosed while they are asymptomatic and relatively contained, or to be cancers that have been growing slowly for a long time (more likely as the 70 year old has been alive longer).

Obviously, I don't know if this is the case or not. If anyone has experience in this field, I'd love to hear about it.

I think it's important to have some conception of a personal "risk tolerance" when it comes to screenings.

Personally, I've had false alarms and dealt with that stress, and I still find that screenings are worthwhile. Having seen family members/friends die of cancer, I don't find the additional and potentially unnecessary discomfort of screenings to be too much of a cost.

At the same time, I'd never judge someone for feeling the opposite, particularly in cases like prostate cancer, wherein most people die "with it, not of it" as they say.

I'm as skeptical of corporations as the next person, but I'm not sure Microsoft is comparable to a literal authoritarian regime.

The segment who opposed the Civil Rights Act, including the senators who filibustered it for nearly 60 days to force a compromised version and the business owners who openly defied it for years in protest, seems like a good place to start.

To be clear, non-trivial WYSIWYG already works.

Wix and Shopify are both publicly traded companies, while Automattic and Squarespace are both planning 2021 IPOs. Wix shares have gone from roughly ~$17 in 2013 to ~$257 today, while Shopify has gone from ~$20 a share in 2016 to ~$1,184 today. Automattic's flagship platform—Wordpress—is responsible for over a third of all sites on the web, and Squarespace was generating over $300 million in revenue back in 2017.

Webflow has roughly zero chance of replacing frontend developers, but the "everyone who needs a website and doesn't want to hire a frontend team" market is pretty large and has been for a longtime.

What makes Webflow interesting to me, and where it is a gamble, is where it stands on the spectrum of developer-friendliness to "can kind of use a computer"-friendliness. They seem to be betting on a change in the market, particularly around the emergence (or rapid growth) of a certain demographic: people who are familiar with web technologies (HTML/CSS/JS etc.) but who don't want to muck around in them directly to build a site—the "know enough to be dangerous" crowd, if I had to give it a name.

I don't know if that will work out exactly, and I personally don't really find it helpful, but it's interesting. At the very least, I personally know a lot of teams building non-SaaS products (infra tools etc.) that use Webflow for their frontend.

Seeing this comment made my morning. As someone who spent way too many years avoiding "hard" subjects that I'd "never understand," who now works in some of the headier waters of computation, I wish more people had this experience of being introduced to a "hard" subject in a way that clicked for them, so they could similarly realize how approachable it is/how capable they are

Best of luck, and keep it up!

@mods - I changed the title from the original "Russian Software Company May Be Entry Point for U.S. Hack" to be more clear, as "Russian Software Company" felt vague and linkbait-y to me. Apologies if this violates the "No editorialized titles" policy.

Big time anecdata incoming:

I've been around a relatively large number of studios (more specifically, people who have founded studios), and from my experience, this post doesn't match up to reality.

From what I've seen, there are categories (often deep technical, bordering on academic niches) where studio-ish approaches work--which is why research labs and similar models have existed for a long time. The studio model, however, doesn't seem to confer any automatic benefit to businesses in general. Most studios I've seen have been full of sort of "meh" SaaS's and zombie startups.

At the same time, I've seen friends have a lot of fun running a studio, even if it didn't lead to a slew of giant exits. One man I worked with had successfully exited his business after 15 for a very large sum, spent a long stretch doing angel investing and some advising, and realized he just loved the early stage environment and helping out founders, even if he didn't want to personally build another startup. The studio he founded, which has been mildly successful but not at the level that someone in VC might aim for, has brought him an immense amount of joy and continues to, and the community that has emerged from it is full of some really amazing people.

So, yeah, I don't believe we're entering a golden age of studios in which products developed within them will dominate every niche--with the exception of niches where studio-like models have always dominated--but I do think studios can still be fantastic in a social/personal sense even if they aren't producing Google's.

Big +1 for tiddlywiki and all of the tools built on it. The single html file really makes it easier to distribute across devices. There's a pretty cool ecosystem built up around it:

https://tiddlywiki.com/#%24%3A%2Fcore%2Fui%2FControlPanel%2F... https://www.reddit.com/r/TiddlyWiki5/wiki/pluginsandresource...

I wonder if there has been a trend away from work that encourages collectivism, and if that is relevant here. In the 50s, for example, over 30% of workers in the US belonged to labor unions. Many, obviously, had also recently served in the military. The public sector had just been expanded radically by the New Deal.

I can imagine that having a tangible cultural impact, particularly around our sense of collective ownership and duty.

I love all these Roam Research style tools--always felt this was a really cool idea.

If anyone is interested in open source options, I'd recommend:

Foam - A Roam-inspired tool for markdown note taking--ahem, personal knowledge management (same features as Organizedly)--that runs on VSCode/Codium. I personally use and love it. Syncs across all my devices with Git, and deploys easily on free-tier Vercel, which is a nice way to check my agenda or read a note on my phone. https://foambubble.github.io/foam/

org-roam, for those emacs/org-mode people. I always wish I had the fortitude to become productive in emacs, but it seems it never will happen. https://github.com/org-roam/org-roam

This is an interesting comparison because I think it actually highlights something deeper here, which is that the kind of shamelessness described in this post is an effective—if not optimal—strategy for a fundamentally different type of reward.

Paris Hilton, for example, was certainly already in a privileged tier of humans—as were/are many others who are born into similar circumstances. It seems, however, that the reward she was optimizing for (building a platform around her persona, effectively turning her life into a consumer brand) was not the same as everyone else in that circle. Her shamelessness seems like a very effective policy, in that context.

Ibra (one of my all-time favorite players) is also a perfect example of this. Despite being a high-level competitor and objectively world-class player, he is also interested in using his personal brand as a monetizable tool in a way that many other top players do not (though, it's fair to mention that players like Maradona have done this historically as well). For example, Ibra's move to the MLS—and the fact that he was paid a record breaking 7.5 million per year—certainly required him to be a great player, but it's also fair to say that the platform he brought with him made him especially attractive to a team looking to sell jerseys and season tickets.

So yes, I think you're right that the table stakes to shamelessness being effective are the essential building blocks of success in one's field—Paris Hilton/Donald Trump have to have wealth and access, Zlatan has to be world-class, etc.—but I think that what that shamelessness buys them is a different kind of success, one they might not achieve without shamelessness (or without some parallel promotional strategy).

To be clear, over 90% of healthcare workers working in hospitals are vaccinated against the flu every year:

https://www.cdc.gov/flu/professionals/healthcareworkers.htm

This is wildly inaccurate.

The data cited in that ABC article, published in 2008, is taken from the 2006-2007 flu season. Over the last 14 years, much has changed.

For example, in 2019-2020, over 80% of healthcare workers received the flu vaccine, including over 90% of those who work in hospitals.

Source: CDC https://www.cdc.gov/flu/professionals/healthcareworkers.htm

If we are comparing the relative safety of medicines, which this article does in its conclusion, then it is the highest bar.

I mean, sure. I grew up poor and without much in the way of access to education, and the internet was really a saving grace for me, in terms of self-education, income, and simply finding community.

At the same time, do you think the average income of a given internet business--let alone the segment that are run by solo entrepreneurs in their basement with just a computer and a $5 droplet--compares positively to the salary of the average CS graduate? Because we're talking about inequality, not asking "can poor people possibly make money online?"

The fact that the internet has made things better does not mean the problem is solved.

It's important to keep in mind when reading anything about GPT-3 that the stated objective in training it wasn't to produce any kind of AGI, but specifically to see if a model could perform context-specific language tasks without first being fine-tuned to the specific context. For example, GPT-2 (GPT-3's predecessor) was originally used in AI Dungeon, but it first had to be fine-tuned on a large corpus of choose-your-own-adventure texts. GPT-3 doesn't need to be fine-tuned—it just works.

From the original paper:

"Here we show that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches. Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous non-sparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks and few-shot demonstrations specified purely via text interaction with the model."

Of course, it makes sense that people critique GPT-3 in terms of its actual progress towards human-like intelligence, since every news publication writes about it like it's Skynet and OpenAI's stated goal is AGI. I do think, however, that we miss all the parts of GPT-3 that are exciting and innovative when we view it through the binary of "Is it really understanding as a human does?" (Not that you were reducing it to that, just speaking about the conversation around GPT-3 more broadly.)

This also isn't to suggest that the education system is the only thing at play here.

I understand what you're saying, but I think there's some "both sides"-ism at play here.

Yes, there is a level of scientific ignorance displayed when people hand wave and claim "the science" is on their side without actually understanding the science. However, there is a huge qualitative difference here, illustrated even by your example of quarantine recommendations.

If we approximate an answer to "What is the optimal quarantine strategy?" through the combined efforts of various sciences--statistics, biology, etc--then while the results are not guaranteed, the recommendation is certainly not unfounded. It may be incomplete, but it is built off of decades of empirical evidence.

Conversely, outright ignoring recommendations about COVID or climate or something similar, simply because "science isn't always right" or because you "aren't a scientist" is a different thing altogether.

I suspect, somewhat anecdotally, that this begins with the educational system (at least in the US). Reflecting on my education, at least, it's remarkable how much of it boiled down to exercises in reading comprehension. Even many of my science classes were really just training my ability to read a passage and respond to a short answer question. There were many "alternative" math tracks offered--home economics you could take home economics after algebra 1, if you weren't a "math person"--and just generally, it was constantly reinforced that math was a specialization, not a fundamental skill.

This is exactly right. The Futurist's love of technology is brought up a lot as a positive aspect, but it's important to contextualize their notion of technology.

From the manifesto: "We will destroy the museums, libraries, academies of every kind, will fight moralism, feminism, every opportunistic or utilitarian cowardice."

The internet is not a technology they would have cared about, and in fact, runs pretty counter to their values. When they say technology, they mean machinery.

Out of curiosity, who do you worry is "pretending this is the path to a successful web business?" Because the author certainly isn't doing so.

If you scroll down just a few paragraphs from the sentence fragment you quoted, the author says:

> "My main goal for the next six months of Tiny Projects is to learn how to get better at making money on the internet from my projects.

A Product Hunt launch can easily get you a spike in traffic, but afterwards I really don't know what I'm doing. There's this whole other level afterwards called "sales & marketing" that I want to master."

The author is extremely explicit throughout the piece that these are not built to become large businesses—hence the constant use modifiers like "tiny" and "micro." He also makes it clear that he treats this as a training routine for getting better at idea generation, shipping products, etc.

The fact that this was posted ten years ago underscores that this is not an HN problem, but a perennial challenge of any large community.

I don't think most people grok exactly how difficult moderation is, and how that difficulty doesn't just scale, but rather evolves fundamentally as a community grows. Moderating a smaller community is sort of like setting a certain tone within your group of friends and figuring out ways to encourage them to follow suit. Moderating huge communities is more like designing a judicial branch.

For what it's worth, I think HN is one of the best examples of successful moderation. The team balances their roles as curators and moderators well, they seem to work hard at intellectual honesty in their policy design, and they do a fantastic job of combatting marketing schemes and bots. No doubt there are areas where I don't entirely enjoy HN, but by and large it is the most consistently enjoyable community of its size for me.

I don't know how much of a role this plays, but even with lower market rates, employing a team in Europe can be an expensive, unwieldy task. Firing in particular seems to be a lengthy, drawn out process. I don't think that's a bad thing, but I can imagine how a significantly increased cost for every bad hire might increase the risk profile of a small startup.

This is the epitome of a straw man argument.

The original commenter was suggesting that incarceration for consuming cannabis is irrational, not that citizens shouldn't be responsible at some level for the collective good of their society.

Additionally, the implied causal chain you've laid out is fallacious. Ignoring the vagueness of what "troubled" means exactly here, if it is "troubled-ness" that makes someone prone to both crime and cannabis use, then clearly it is the troubled-ness, not the cannabis use, that causes the criminality. And if the goal is to reduce criminality, as is implied in your post, then whatever causes people to become troubled must be addressed—which, again, isn't cannabis, as it seems being troubled is a prerequisite to using it.

This is akin to the classic example of banning ice cream to reduce the murder rate. Ice cream sales and murder rates increase in a correlated manner—but only because both increase in the summer.

Also, as a final point, the fact that drugs like cannabis are criminalized turns any connection between cannabis and crime into a self-fulfilling prophecy. The real societal cost to the criminalization of cannabis comes via the violence and destruction associated with any criminal marketplace. The last time we saw massive organized violence between criminal organizations over alcohol, for instance, was the last time it was criminalized—i.e. during prohibition.

But yeah, something something libertarians something something cultural decay.