fro0116

Yeah, Mercury's going to have to up their game if they want to stay relevant. I remember my disappointment when I found out I had to park 250k in a 0-yield checking account to qualify for the 1.5% yield on a savings account that can only do something like 5 transfers a month, making it a complete non-starter for running a real business. It felt like they're playing all the same old tricks on us as the incumbent banks, just with a prettier UI.

This new Brex account with 0 fees and 1.6% yield on the entire deposit with no transaction limits feels like a breath of fresh air in comparison.

I wouldn't be so sure, there was the rather plausible speculation that the 3% from Robinhood was a loss leader that wasn't meant to last long, in order to drum up hype for the launch. They could very well have been planning to earn the fed rate and fronting the difference for a period of time and then bait & switching (and/or introducing some premium plan to make up the difference).

Also, money market funds, which are generally considered cash-equivalents, don't usually yield north 3% either afaik, and I can imagine putting customers deposits in riskier higher yielding investments could have also have ran foul of banking regulations.

Brex is using neither of these strategies here as far as I can tell, and they're providing a decent amount of transparency into how the funds are going to be handled, so it should be fairly easy for regulators to review. If regulators do take issue with their strategy, I suspect we'll probably hear about it in the news soon enough. In the mean time, it's still an early access product, so it's not like we can use it immediately anyways (I'll definitely be signing up though).

From the first sentence under the "What are securities" section:

> SIPC protects stocks, bonds, Treasury securities, certificates of deposit, mutual funds, money market mutual funds and certain other investments as "securities."

Surely I can't be the only one irked by the fact that they're bringing the concept of "copies" and "borrowing" to what are just bytes sitting on hard drives that can be copied and transmitted at 0 marginal cost?

I agree that it's a better user experience than the non-digital alternative, but I'd like to see digital books at libraries be available for free to anyone who wants it at any time, without any arbitrary restrictions whatsoever.

IANAL, but according to this: https://www.sipc.org/for-investors/what-sipc-protects

> SIPC protects against the loss of cash and securities – such as stocks and bonds – held by a customer at a financially-troubled SIPC-member brokerage firm.

It sounds like SIPC protects against loss of cash for the purpose of purchasing securities _and_ the loss of the securities themselves. So in this case it sounds like by sweeping customers' cash into a money market fund, SIPC protection would apply to the money market fund investment.

I think Robinhood's issue might have been that they didn't actually do any securities trading and tried to just rely on SIPC insurance for customers' cash that's just sitting around.

Regardless of if my interpretation is correct, I'd be _really_ surprised if they just somehow ignored the whole Robinhood fiasco and managed to make the exact same mistake on the regulatory side.

From the deafening silence I'm going to take the less charitable interpretation that it's meant to enable Cloudflare to essentially sell Warp users' IPs to Cloudflare customers as an added perk.

I wished on more than 1 occasion that I could just use one of my ZeroTier nodes in other locations as an exit node and tunnel traffic through it though.

I think that might already be possible but it'd be nice if the configuration was abstracted away behind some simple interface.

That'd allow it to replace Wireguard and Mullvad for me as I mostly use it as a means to bypass georestrictions.

Wow... this is actually quite tempting for $595!

Are there any limits to how much you can make use of this offer or the kinds of spaces available?

_Surely_ they won't let you just camp an entire private office in SF all year round when it's supposed to cost upwards of $1000 per month?

They have an app now, works much better than the text receipt feature in my experience so far: https://brex.com/mobile/

Brex also has a Ecommerce card that's based on projected monthly sales: https://brex.com/ecommerce/

That might be more directly comparable to the Stripe card.

Most successful companies do both. Listening to your users is of course important, but it can't be a wholesale replacement for data driven aggregate behavior analysis because people have their own biases can forget things (and sometimes lie) and individual feedback can't be easily generalized to your entire population of users past a certain scale.

> forgetting the kind of person who is techie enough to use RSS feeds is probably techie enough to turn off telemetry

More like they have no choice but to ignore them because these are users who explicitly chose to be ignored.

If you want your usage patterns to affect product decisions in a product you use, then don't disable telemetry for that product. You can't have it both ways.

FWIW, there are very few companies who go to as much lengths as Mozilla to optimize for user privacy in their approach to telemetry: https://hacks.mozilla.org/2018/10/testing-privacy-preserving...

That explains why everybody puts up with it. It doesn't explain why a startup like Mercury can't come along and shake up the status quo by not nickle and diming their customers like the greedy fucks that are the incumbents.

Missed opportunity IMO.

Interested in learning more. Could you provide some sources on why Qualcomm Atheros is "Tier 1"?

> I can totally understand it in cases where the alleged offense is something like uploading copyrighted content to YouTube, where there is clear evidence and an audit trail saying "Here's what you uploaded, here's when you uploaded it, and here's the point where the rights holder registered it in ContentID, etc"

Probably not the best example. There are countless reports of ContentID falsely claiming copyright violations. I personally had a gaming video muted for violating some copyright by some company I never heard of when the only thing playing were ingame sound effects (no music).

I'm curious what the article is referring to when it comes to collision resistance. I was looking at the GitHub repo posted below and saw only claims of "strong" collision resistance: https://github.com/google/highwayhash

Unfortunately I'm not familiar enough with the details of the collision resistance properties of the other hash functions here to know how they compare. Could someone elaborate on the specifics of why HighwayHash compares poorly in terms of collision resistance to the others in the benchmark?

I use Google Voice extensively too. The problem I often run into is a bunch of services (like Uber, and one of my credit card companies Synchrony Bank) don't accept Google Voice numbers. What do folks usually do in those cases? Are there services like google voice but don't have the non-acceptance problem?

I can absolutely imagine why existing tech companies would be interested if the protocol became so popular that they'd be stupid to ignore it and start from scratch with their own protocol, like XMPP was at one point.

I'd be more interested in seeing how they plan on preventing the whole Embrace, Extend, Extinguish thing pretty much every company pulled with their initially XMPP based chat apps that gained market share, turning them into back into closed silos.

Being somewhat of a functional programming zealot (coming from the Clojure camp, so please excuse me if these are stupid questions, considering my lack of familiarity with typed functional programming in general), this is something that has always bothered me about TypeScript's type system, the poor support for higher order functions.

For a prime example of this, take a look at the type definition for the ramda pipe function: https://github.com/DefinitelyTyped/DefinitelyTyped/blob/mast...

It's a hard coded list of 10 signature overrides that allows it to support up to 10 piped functions. Obviously the actual ramda pipe function can support an arbitrary number of piped functions at runtime, but the types as written only supports 10.

This seems awfully inelegant and inflexible. I assume there is some fundamental deficiency in TypeScript's type system that forces people to specify types this way, otherwise it would have been rewritten by now.

Is higher kinded types what's missing to be able to express higher order functions and functional composition in an elegant way? If so, can someone provide an example of what the type signature of a pipe function would look like with higher kinded types? Any ideas if higher kinded types are being considered as additions in future versions of TypeScript, or if that's even feasible at all?

Thank you for providing your side of the story. While I can no longer edit my post to point to this, I have upvoted this post to hopefully bring some visibility to it, and encourage others to do the same.

Until someone can provide a chat log of what was actually said in that conversation that led up to that comment, it's still going to be difficult for any outside observer to make up their own mind on whether or not those words were indeed taken out of context to the degree that you suggest. But I can appreciate that it could be frustrating to deal with someone who behaves as you claim, even if his intentions might be to keep the protocol open and secure.

To quote one of the follow up comments from the author:

> If you go to matrix.org and look at the list of about a dozen or so servers: you will find that none of them actually work except the reference implementation, and maybe sometimes Construct. Even thus, the phrase "able to build" is questionable. I have spent months reverse-engineering their software and its interactions before, and after, it was at all documented in this so-called standard (by the way, it's just documentation of their software -- errata and all (and rather poor)).

> Construct server is the single survivor out of the ones listed and even more who have attempted and given up early which we don't know about. That being said, it is still incomplete.

A spec that "Anyone can implement" doesn't have much value if it's so bad/incomplete that in practice nobody else is actually able to create a complete working implementation of it as the post claims.

And regardless of what they're planning on doing about any security issue, there's never a good reason for a discussion of a potential security issue in an "open" protocol to end in "good luck talking to your own federation." That's the behavior I'm pointing to that's hostile to competing server implementations.

Of course, the quote could be taken out of context, or straight up made up. But until proven otherwise, I'm willing to give the benefit of the doubt to the small independent developer who seems to genuinely care enough about the openness of the protocol to build their own server implementation.

I had a pretty positive view of Matrix until I came across this post and the follow up responses from the developer behind an unofficial Matrix server implementation: https://news.ycombinator.com/item?id=19365968

I'd say tread carefully given their apparent hostility towards competing server implementations, which is literally the only thing that makes the protocol meaningfully "federated" to begin with.

This part in particular was deeply disturbing:

> I can quote the CEO of new vector in an argument we had about the insecurities of the protocol and what needs to be done to fix them where he said "good luck talking to your own federation." That reveals a lot.

Not exactly the kind of attitude I'd like to see from the stewards of an "open" protocol.

I'm a huge fan of the WebTorrent protocol too, and think it has potential to completely transform the content distribution paradigm on the web.

However, the JS implementation leaves much to be desired. It's a huge resource hog compared to native libraries like libtorrent and torrents often freeze for no apparent reason on faster connections. Development seems to have also stalled quite a bit in recent times.

At this point I'd be more confident betting on something like IPFS in the long term.

That's one way to see it, and it's certainly not unreasonable to see it that way.

But looking at it another perspective, you can also argue that making MMOs with persistent worlds that can only be supported by subscription has become so risky of an investment that nobody is willing to experiment with drastically new paradigms/mechanics because historically an overwhelming majority of those who tried have failed, which could be what led to the creatively bankrupt landscape we have today.

It's impossible to know which perspective is more accurate because it's clearly a chicken-or-egg situation. But one thing that we can observe today is that practically (or maybe actually?) nobody even tries to make MMOs with subscriptions anymore, and new MMOs today end up defaulting to freemium (and all the morally bankrupt behavior that the model leads to), whatever the reason for that might be, and I think that's a shame.

> World of Warcraft, Final Fantasy XI, Final Fantasy XIV, and EVE Online all still successfully use the subscription model (with varying levels of free trials).

The most recently released game in that list is Final Fantasy XIV, which was released in 2010, almost a decade ago, and the rest are about a decade older than that. A _lot_ more MMOs since then have tried the subscription model and failed. It may have worked in a few rare cases in the past, and some of those might still live on today through sheer inertia, but I think it's reasonable to claim that subscription is generally no longer considered a viable model for new MMOs.

I'm not saying some publishers won't find ways to extract more money from players by adding freemium features on top of subscription games if they could. Of course some would, greed knows no bounds. But I am saying that they can't afford to charge a subscription fee to begin with, if they want their game to become successful in today's climate, even if they intend to use that as their only source of funding, forgoing freemium features entirely.

> I don't find it difficult to fault exploitative behavior just because it's profitable.

I think I made it pretty clear I find those practices despicable too. But if your only option is the freemium model, you don't really have much of a choice but to discriminate based on who's willing to pay and how much. That's literally the only way the model can work, by definition.

> The designers of free-to-play games, by using an intermittent variable to dole out small prizes, found that they could keep players engaged—and spending—for longer.

I'm really glad the article also touches on the loot-box phenomenon. I think it's by far the worst thing to ever happen to game design.

It's exploitative in the same way that gambling in general is exploitative, but isn't bound by the tight legal restrictions that prevent gambling companies from targeting children who might be most vulnerable to these exploitative tactics and more likely to fall into a vicious cycle of addiction early in life, which can have dramatic, literally life-changing consequences.

That said, I can't really deny either that at least part of this trend is self-inflicted on the part of the gaming community at large. Gamers have all but collectively rejected the subscription model where we pay for the games we play on a recurring basis. For games like MMOs that require ongoing funding to develop new content in order for the community around it to thrive, what other funding models outside of the freemium model do designers really have?

The DLC/expansion pack model also worked adequately for some time, and still does for games that involve mostly episodic experiences, but not so much for games that have living worlds and require ongoing development of new content and incur ongoing infrastructure costs. It also comes with similar problems as the model of paying for software upgrades, because developer incentives and users' needs can become misaligned. Oftentimes in this model, bug fixes and quality of life improvements get sidelined in favor of more content/features, and the community becomes fractured across expansion packs, undermining the natural network effects that these kinds of games inherently tend to create.

It's also really difficult to fault freemium games for using exploitative tactics that target players who do end up paying to pay more, when the distribution of people who actually pay for things in these games is often so wildly unevenly distributed that the top percentage of paying players often account for a disproportionate percentage of overall revenue, while the vast majority of overall player-base pay absolutely nothing, and those who do so often wear it as badge of pride, no less.

I've always wanted to build a game myself, but I could never figure out how to align my own need for ongoing revenue to support the development of a game that I might love and end up wanting to keep working on indefinitely, with that of a gaming community that has soundly rejected any funding model that makes it possible to achieve that without resorting to tactics I'd rather not stoop to.

This reminded me of an IndieHackers episode from a few weeks ago, where they interviewed the founder of https://pioneer.app/, which attempts to use gaming mechanics to motivate people to work on interesting projects and be more productive. It's a really cool idea that has the potential to do a lot of good for the world imo, so I have hope that this _can_ definitely be done.

Highly recommend listening to the whole thing if you're interested in this stuff: https://www.indiehackers.com/podcast/080-daniel-gross-of-pio...

Well said. Reminds me of something I heard in some interview about what makes the valley special that really stuck with me. It goes something along the lines of:

"There's no other place quite like the valley for starting and scaling a company from the ground up. No other place in the world can even come close to the valley in terms of the sheer number and concentration of engineers, leaders, and investors who have experienced the scaling of a company from the ground up first hand. And those are the people you want by your side when starting your own company, because they can apply their learnings to help your company grow at an accelerated pace and avoid the pitfalls they encountered along the way."

> user configurable keyboard shortcuts

This is one of those things that's sorely missing for me. Firefox has keyboard shortcuts for most things I want to do in it, but they're often really inconvenient and I'd like to be able to change them to match those in other tools I often use.

Some other things I'd love to see in a browser:

- A command palette to make functionality & shortcuts more discoverable

- Fuzzy tab search and switcher

- Customizable split panes

On the whole I think browsers could benefit a lot by learning from UX patterns in modern editors like Atom/VSCode.

Another really cool project that makes use of CRDTs for collaborative editing is xray from the folks who worked on Atom:

https://github.com/atom/xray#text-is-stored-in-a-copy-on-wri...