alsargent

San Francisco Bay racer here. Very nice website — thank you. A few requests:

1) To make it easier to identify these buoys, please take the information in the “characteristic height” and “structure” columns and associate them with each buoy. Expand the brief description in the characteristic height column so that it’s understandable to a layperson, eg “Fl Red 3” should be “Flashing red light every 3 seconds”.

2) What would be a good way to add non-USCG navigational aids, eg racing buoys like Blackaller/Crissey, Fort Mason, Yellow Bluff/Easom, etc.?

Agreed

Interesting stuff

Thanks. As an Evernote user since 2007, and paying customer for several years, this sucks.

I’m curious: why is Notion not on your short list?

Maybe this is a dumb question... but why doesn't Reddit lower its API prices to a point that Apollo and other app developers can afford?

Plenty of other companies have figured out how to price APIs in a way that works for developers: AWS, Twilio, Stripe, Okta, MongoDB, and Plaid, to name a few.

It's not like these companies aren't making money with their API pricing; they've all generated enough in revenues and profits to drive their valuations into multiple billions of dollars.

It's as if Reddit didn't do the basic work of rolling out API pricing: talk to customers, find price points they can live with, offer prioritized customer support in exchange for API charges, etc. Literally hundreds of software companies have followed this playbook, and have rolled out API prices without drama.

Am I missing something here?

Will be popular with shoplifters, package thieves, and bus thugs.

I wonder if note taking applications like Notion (or new startups) might include something this as a capability to improve their search and summarization.

Thank you! (ArangoDB employee here...)

“The brevity of our lives breeds a kind of temporal parochialism—an ignorance of or an indifference to those planetary gears which turn more slowly than our own.”

… great point.

Here are some trends that could drive a fundamental shift in social media:

1) Trust is the scarcest element in social media today. Any social media company that is built on advertising will never have the trust of a subscription-based social media company. Companies that address scarcity tend to be successful.

2) What's no longer scarce: the underpinning technologies of social media: capturing and displaying photos and videos on multiple types of devices, recommending new social connections and posts. What was cutting-edge in 2004 is now well-known.

3) Meanwhile, users are getting increasing used to paying for subscriptions: app stores, streaming services, SaaS applications, cloud services, etc.

4) Connecting socially with others is a basic human need. This only increases as some kinds of jobs can be done from anywhere, and friends relocate far away.

5) As Facebook/Meta and others pursue the novelty-driven user experience of TikTok -- "show me what's interesting from people I don't know" -- it creates room for companies that want to get back to meeting the need for keeping in touch with friends and family, even when remote.

6) Large tech fortunes have created a donor class focused on legacy, not profit. Example: MacKenzie Scott, the ex-wife of Jeff Bezos. Or Craig Newmark, of Craigslist.

--

Put all these together, and it seems like new social media companies could be created along the following lines:

1) Mission-focused. Focus on social connection first, not whatever drives the most revenue. In other words, don't get pulled into the latest fads, as Facebook is doing with TikTok.

2) Subscription business model. This eliminates the conflicts of interest that drives Facebook's trust-eroding privacy practices. Again -- trust is the scarcest element.

3) Subscriber-owned business. Each subscriber owns a portion of the company, and thus the company has a fiduciary, legal obligation to protect their interests. This is similar to what Vanguard does -- investors each own a portion of the company -- which forces Vanguard to act in their interests. It's the opposite of Facebook/Meta's ownership structure, where Mark Zuckerberg controls 90% of class B shares, giving him control over the company. [1]

4) To fix the cold-start problem [2] inherent in building a business with network effects, make the service free until it gets to a critical mass of subscribers. We can debate if critical mass is 10 million users, 100M, 1B, or some other measure. But be transparent about the threshold, and the subscription price once its hit. Speaking of price...

5) Keep entry level prices low to be point of being negligible for the vast majority of users. Maybe one dollar a month. Whatever it is, keep it lower than most other subscription services in order to encourage adoption, but not to shift back to the problematic ad-driven model.

6) A very low subscription price, at scale, can fund innovation. 100M users at $1/month is $1.2 billion per year. That's enough to pay cloud infrastructure and the engineers to build and run apps. Back-of-the-envelope path: suppose for argument's sake that half of that, $600M, goes towards cloud service providers. That's approaching the $1B/year that Netflix spends. The other $600M could fund 2000 engineers at $300k/year/engineer. That's enough to build a great deal of capabilities and bring them to emerging platforms (like AR glasses, cars, IoT/smart home...).

7) A business like this probably might not attract traditional venture capital funding. Even if every one of Facebook's 3 billion users all switched to this business and paid 1 USD/month, that would be $36B per year. That's well short of Facebook's $120B/year [5]. Who might fund it? A set of mission-driven investors, who wants their legacy to include a trusted, self-sustaining organization that socially connects the world. Craig Newmark could be one such investor (at least advisor), having built one such Internet institution (Craigslist) that facilitates community and commerce in an economically-sustaining manner. But there could be many other investors as well. Again, the technologic acumen and capital required aren't what's scarce; trust is.

[1] https://www.morningstar.com/articles/1061237/how-facebook-si...

[2] https://www.amazon.com/Cold-Start-Problem-Andrew-Chen/dp/006...

[3] https://www.linkedin.com/pulse/netflix-pays-1-billionyear-am...

[4] https://datareportal.com/essential-facebook-stats

[5] https://www.statista.com/statistics/268604/annual-revenue-of...

To get a sense of how many solar panels and batteries we'll need, drive around Houston, Texas some time. You'll see mile after mile of chemical refineries. Granted, some of these are for plastics and not energy production. But a lot ARE for energy production -- and it's crazy to imagine how many square miles of solar and battery factories we'll need to provide an equivalent amount of energy production.

Downhound creator here. Thanks for linking to the site; hope it was able to provide a bit of clarity on the situation on Google yesterday. Google's outage was pretty short, around 10-15 minutes or so.

What new features/data would you like to see in Downhound in the future?

Thanks.

I hear your point about the last sentence. If I read "Glory to America", I'd more than cringe.

However, the last sentence in Roman characters is "Slava Ukraniya" which has emerged as more of a call to defend Ukraine than jingoism against outsiders. The same people that say this are also giving food to Russians that surrender and letting them call their parents.

If you have a website, you can support Ukraine right now.

Not all Russian citizens have the information about what their armed forces are doing in Ukraine, and what casualties they've already experienced.

Add the script in the link to your website to show a popup for visitors with Russian IP addresses with links to truthful news sources and information about prisoners of war.

"We never spent a dime on marketing."

... except for Luis' salary, hosting the website, and the rest of the marketing stack, which might include Calendly, video conferencing, mailing list, etc.

(InfluxDB employee here. Sorry I should have mentioned that in my original post.)

Migration is something we feel is very important, which is why we've delivered a single-CLI-command upgrade from 1.x to 2.0. Here's the documentation: https://docs.influxdata.com/influxdb/v2.0/reference/upgradin...

Also, screens last a long time, but it's the computers themselves that need continual ugrading. Case in point: we have a Dell screen from 18 year ago. The Dell desktop long since became obsolete.

So this makes a lot of sense economically, especially for cash-strapped organizations like schools that have to manage large fleets of machines.

Stuff that isn't touched and lasts long (screens) don't get replaced. Stuff that is touched (and likely to wear out over time) like keyboards or gets out of date due to Moore's Law (computer) is fused into one package, so there are fewer cables to worry about getting unplugged, worn out, etc.

I'll make two bets:

1) PC makers start coming out with their own keyboard computers with a similar form function within the next 12 months.

2) A future Mac Mini has this same keyboard form factor. (Perhaps with a "keyboard + magic trackpad" form factor.)

Maybe I missing something, but I've published several tutorials on Medium, all outside their paywall, and all appearing in google search results. It's a nice blogging platform. Not sure what the issue is.

For instance, this article I wrote is still available outside their paywall: https://medium.com/@alsargent/monitoring-aws-with-telegraf-i...

Consider Sales Engineering. Often in tech companies, there are two types of salespeople: nontechnical folks, and sales engineers.

The nontechnical folks have titles like Account Exec, Sales Development, Inside Sales, Regional Manager. These are the "people people" in sales. Extroverted, greater social skills. Your classic salesperson.

But software is technical, and often gets sold to technical folks. At some point in the sales cycle, a company's sales team needs to talk "brass tacks" and show what the software can do. This means technical presentations, demos, pilot projects. It will be more at infrastructure companies like, say Snowflake, and less at SaaS companies like Salesforce.

For these technical jobs, you need to have sales engineers. You don't need to be a great developer. Instead, you need to enjoy learning technology on your own, and able to "geek out" with fellow geeks. Given your work as a hobbyist, it seems like you definitely have the first attribute.

Hope this helps.

Hacker Newsletter. The best* of Hacker News, delivered once a week. Each newsletter contains a few useful articles that have helped me in work and life.

* not sure how they figure out “best” but the articles are general good.

Interesting story about how this developer got called by the NSA, who instructed him to hang up then call back the publicly listed number for the agency. Exactly the steps that big banks should take. https://medium.com/datadriveninvestor/why-the-nsa-called-me-...

Test Automation U. just released a new online course today: Introduction to Capybara. Here's the link:

https://testautomationu.applitools.com/capybara-ruby/

For those not familiar, Capybara works with Ruby and is a domain-specific language (DSL) that comes with built-in methods that allow you to visit a page, fill in a textbox, click a button, click a link, check a checkbox, choose a radio button, or select an item from a dropdown and search within a section of the page.

The instructor is T.J. Maher. You can read more about his background here:

http://www.tjmaher.com/

Hope you like it!

(Disclosure: my company sponsored the creation of this course, and has a small link to our site at the bottom. That said, Capybara is free open-source software that is unconnected to my company. The instructor is not an employee of my company.)