cvwright

According to the last 107 years of history, yes.

Making $10M ARR companies pay for the software that they use is not enshittification.

BSL code automatically converts to open source at a specified date. So probably several releases since then are now as open source as anything else in the world. And if not, then they will be soon - BSL allows a maximum 5 year delay.

I have about 30k lines of Swift in my app + Matrix sdk, all heavily using async/await. Plus another couple kloc in a Vapor API backend.

Swift async/await is quite nice.

The Actor stuff is IMO way oversold - reentrancy issues make it not actually useful for the things that they claim, like querying a database. It’s still a useful construct to have, you just need to be aware of what it does or doesn’t provide.

True, but we are living through an obesity epidemic, and it’s not just the rich people who are fat.

Ah, yes you do have to run a service of some kind to interface with Apple. For example the Matrix project provides Sygnal https://github.com/matrix-org/sygnal which the owner of a Matrix client can run on their own systems.

It's not super resource-intensive if you run it yourself, but I can imagine how a "buy vs build" decision for other apps could easily lead to a $$$ monthly subscription with a specialized 3rd party.

These prices make sense when you consider that the bulk of the cost of providing Zulip to you is not the cost of the compute and storage and bandwidth - those are fairly minimal.

It’s the cost of the developers who fix bugs and keep the product moving forward. Those cost a lot more, and you need them regardless of where your instance is hosted.

I don’t think Apple charges for APNS do they?

I have 500-600 users on my app and have never paid. Maybe it’s different at scale?

Also, that’s about half of what you’d pay for 11 users on most other group chat services.

From what I can tell, the average is about $8/user/month. And no, I don’t understand it. For my former employer, a large public university, these services would cost about $240k/month before any special discounts. For that price you could hire a whole team to build and run your own product. Make it make sense.

Your story supports the position that it's safer to take the FAANG job.

This guy found an underserved market opportunity, built something, got some success marketing it, and made a whopping total of $40k.

Meanwhile the FAANG guys are making that every two months, without all the risk.

That's not bad advice - you just managed to overcome a lot of real obstacles. Humble brag.

The community has been trained over many years to believe that anything FOSS is immediately trustworthy, and that anything non-FOSS is immediately untrustworthy.

This gets exploited all the time when companies use FOSS licenses to get traction, then pull the rug and switch to a more restrictive license once customers are locked in.

On the flip side, if a company tries to be honest from the beginning by using BSL (or similar) they get endless grief from religious zealots.

AGPL is the magic unicorn license that is just restrictive enough to protect you from your competitors, while still passing the FOSS religious purity test.

Many of them are going AGPL, including the one we're discussing. Also recently Minio and the Synapse Matrix server from Element.

It's not necessarily about being better. Sometimes a competitor can beat you just by being cheaper, or by being better connected.

One example is that the Matrix project recently took a budget hit when the core team apparently lost a big bid to a "large system integrator" who seems to have used their own open source code to bid against them.

Or give the money to a more “working class” med school that trains regular doctors to work in regular practices and hospitals.

Hopkins is the elite - those doctors are going to be the best of the best, and they’re going to be well connected and compensated regardless of where they started.

So true! Copyrights that last 20 years would be completely reasonable. Maybe with exponentially increasing fees for successive renewals, for super valuable properties like Disney movies.

Lately it seems these are always revised down after the initial release. In a month we may find out that the real numbers are even worse.

Spring 2022, when spaces were coming out of beta. My boss tried to switch spaces and got a completely blank white screen. Not a good look.

Personally I use Element every day to keep in touch with the Matrix community, and it works fine for my minimal needs. It’s better now than it was then.

Hey, thanks for trying Circles. We are doing a lot of work on improving the usability this summer, so yes please send us any feedback.

If you’re a Matrix user, come chat in #circles:futo.org.

Otherwise the Github repos all have issues enabled so please feel free to file things there.

Unfortunately the usability of the available Matrix clients leaves a lot to be desired. When we evaluated Element I think it was in a particularly buggy state.

In contrast, Zulip is fast, simple, and “just works”.

I think, rather than looking at the Source First license as something that’s “a replacement for OSI Open Source but less free” it’s more useful to see it as “a replacement for closed source commercial licenses but radically more open”.

Open source has created some really great backend software and developer tooling, but somehow it’s not great at making nice consumer-friendly client side apps. When you look at the best apps, even the ones like Chrome that use open licenses, they are created by a relatively closed group of developers funded by a for-profit company.

The Source First movement is aiming to provide a way for that kind of closed focused team to be more open with their users. I don’t imagine that many Source First projects are going to be looking for lots of community contributions, and that’s ok - the closed source apps don’t do this either and they seem to get along fine.

You can backronym all kinds of things into it.

But in reality it’s four letters that sound good together. (And for which a domain name was available, I assume.)

Not true at all. We do use the "Source First" license for a lot of things, but all of my projects are OSI Certified (TM) Open Source. For example:

* https://github.com/futo-org/circles-android (AGPL)

* https://github.com/futo-org/circles-ios (AGPL)

* https://github.com/futo-org/matrix.swift (Apache 2.0)

There are a few other little things here and there, mostly Apache 2.0, but the above are the bigger ones.

Edit: There is also Alex's Whisper finetuning repo https://github.com/futo-org/whisper-acft (MIT)

libjpeg 6 is pretty good. It's oldschool C so it's pretty spartan, but it's easy enough to read it and follow what it's doing if you have a basic understanding of the concepts. (Discrete cosine transform, Huffman coding, etc)

And it's only had a couple of vulnerabilities, in nearly 30 years of being on the Internet. That's not quite like DJB code, but it's darn close.

My understanding is that it was written by a very small number of people who were experts in the field, as part of the standards process way back in the early 90s.

We had to contact them in order to add team members. Which still put them in the top 2 or 3.

But a lot easier to get started and do simple things.

The "dedicated server" / "private cloud" market is a garbage fire of manual one-off hackery where you have to talk to a human in order to do relatively simple things that are a couple of clicks on even "2nd tier" VPS platforms. Want to increase the limit for the number of servers on your account? Call us. Want to add a new employee to your team? Call us.

Earlier this year, my team evaluated about 10 different providers, including PhoenixNAP, and only maybe 2-3 of them allowed us to (1) create an account, (2) add team members, and (3) deploy a server without manual intervention. And often these simple tasks required multiple rounds of manual debugging with the support team.

I spent 8.5 years teaching at a lower-tier, public "access" university in the US. Don't get me wrong, it was great for what it did, and I was fortunate to work with a lot of awesome people. But we were far from an elite, exclusive institution like Harvard or MIT.

For many years I served on the CS department's graduate admissions committee. Lots of our MS applicants talked about working on sites for major US / western brands, and lots of those kids did not make the cutoff for even our relatively low bar for admission.

I think about that a lot whenever I see that a multibillion dollar multinational corporation has a web page that doesn't work at all.

They also get paid when you send traffic to free, ad-supported sites full of trackers.

I’m not sure what your point is.

There were worries after 9/11 that terrorists were using stego to plot attacks, posting their messages “hidden in plain sight” inside images on public websites.

Someone (Niels Provos?) did a pretty thorough search and analysis of images on eBay and came up with nothing. Apparently it was just post-9/11 paranoia.

Wow that brought back some memories! I knew I had gone to a pizza place with an animatronic band, and it definitely wasn’t Chuck E Cheese. I’ve never been able to figure out what it was.

Yup, it was Showbiz and the Rock-afire Explosion. https://en.m.wikipedia.org/wiki/The_Rock-afire_Explosion

Thank you!