joisig

As an Icelander, I approve of the name, which means "wisdom" in Icelandic.

Collaborative architectural design is what Arkio does, one of the launch partners.

> But if you decide to open source something, keep your promises, no matter what.

Presumably all older versions of the software that were Apache licensed remain so. They are switching to BSL (closed, but source available and each revision switches to Apache licensed 3 years later) going forward.

I don't see how this is worse than keeping everything closed source from the start, which is your suggestion? Shouldn't we be thankful for all the work they've done so far for free, and if somebody doesn't like the new licensing model, then they can fork it from the last Apache-licensed version and keep going from there?

Take a look at Oban, it very well might fit the bill.

https://github.com/sorentwo/oban

The purchase has not yet gone through, Ardian are saying they cannot continue without changes to the agreement, I believe this is due to some requirements by the anti-monopoly authorities [0].

Míla is the company being sold, it was spun out of the national monopoly a long time ago to provide base layer services to all operators equally but was initially (and still mostly, I'm hazy on the details) owned by the old monopoly. The old monopoly has also been privatized a long time ago but has significant competition.

I agree selling the infrastructure part to PE is almost 100% guaranteed to give bad results for the nation. It's like people never stopped to look at what the typical PE playbook is. But I guess it's harder to stop when everything is privatized already and in the hands of very many shareholders, each of them acting in their own short-term best interest.

Source: I'm Icelandic.

[0] (in Icelandic) https://www.mbl.is/vidskipti/frettir/2022/07/17/ardian_saett...

Takes me straight to a log-in pge with no additional context, is that intentional?

Very cool project! I’m curious how the magician would typically use this, would the scanner and an output monitor be hidden somewhere on stage or similar?

Not the GP but they might be referring to [0] or one of several other articles you will find if you Google "handle a million connections in *".

Realistically you also usually need to perform some non-trivial work from time to time for some non-trivial portion of those connections, which will further load your server, but still.

[0] https://phoenixframework.org/blog/the-road-to-2-million-webs...

This is very cool!

An idea for going further still would be a blog, including a content management system, that is a single, self-contained, self-modifying binary. Write a new blog post, save it and the executable modifies itself to contain the new post and be able to output it.

Postico is a really solid piece of software. Very user friendly and approachable, it's a tool I use every day for working with databases. Solidly recommended.

The original make was also a one-person project, it was created by Stuart Feldman

Thank you for the Manifest V3 critique, the examples you give make it really clear how many things are regressing with this upcoming change :/

I love using the Bear note taking app and am building a Bear-powered CMS so I can run my blog on Bear (with a tag controlling what is published publicly) and also have whatever set of notes I like (again based on tags) available to me and possibly a few others like my wife, behind a log-in, via a browser, as Bear itself is macOS/iOS only.

Success will be when my blog is hosted on it and I open source the code so thst it might be useful to other Bear aficionados. No expectation of anything more.

I use the latest iPad Air similarly. Got it and an Apple Pencil as a more-capable replacement for my reMarkable tablet and my Kindle, turned off all notifications and did not install any distractions. It's a great device. For me the main apps are Bear and Concepts.

Vagrant [1] is quite helpful for this, it lets you easily manage (and check in to source control) a VM configuration with the environment needed for development. Write your provisioning script to set up everything you need to build that particular project, share local directories to the virtual machine, use your existing editor on your main machine.

[1] https://www.vagrantup.com/

Quuu Promote is a kind of content promotion on autopilot, but then first you’d need some content, could hire a content writer on Upwork or via an agency for that.

I've tried probably a couple of dozen methods, paper being a solid thing I go back to every couple of years. Complice (a web app) was interesting.

For the last couple of years, I've used free-format text in a new note in Bear app every day, an ever-evolving template (I generally copy the previous days' document, remove done items and stuff that's no longer important, and plan from there). I do something similar for weekly and quarterly plans, all free-format in my notes app of choice.

I built GRIT [1], originally for use in the Google Desktop project as a way to stop doing nightmare three-way merges of translated Windows resource files. It's been used in Chromium and various other projects since then to aid with i18n, and many other folks have contributed along the way. It is open source [2]

[1] https://www.chromium.org/developers/tools-we-use-in-chromium... [2] https://chromium.googlesource.com/chromium/src/tools/grit/

Back in 2009 or so, I and a few others ran a Google-internal April Fool's prank based on this.

We auto-generated a humongous pull request for the Chromium source code and sent it out for review by all the various source code "owners" (super reviewers) who would need to sign off.

The change used the alternate encoding for curly and square brackets, claiming the original characters were inaccessible on an Icelandic keyboard layout (which is half true, and there were quite a few of us Icelanders around) and that therefore, to be inclusive, we would like to standardize on the alternative representation.

Good fun, although I don't remember if a single one of the reviewers was fooled :)

It's funny because it's true :D

Take a look at https://github.com/elixir-nx/nx

I listened to a very interesting podcast episode [1] on the Thinking Elixir Podcast on this subject recently, an interview with John Högberg and Lukas Larsson. Not covered in this blog post but discussed in some detail in the podcast is that because you get standard debugging symbols from the JIT, you will now be able to use gdb and prof and similar tools when working with the BEAM.

[1] https://podcasts.google.com/?feed=aHR0cHM6Ly90aGlua2luZ2VsaX...

I sometimes wonder why app makers don't "unionize", to collectively threaten to boycott the Apple ecosystem, to pressure Apple into reducing its cut. An iPhone without apps is much less valuable than an iPhone with apps.

Thanks, I will update the post to talk about first-to-file since that changes the tradeoff one is making by deciding whether to file patents or not.

This is true and is the reason I break my own rule in this case - because of large companies that are only willing to pay through an annual PO/invoice process. None of our customers are quite big enough to require the use of a preferred reseller, but I've heard of that arrangement as well.

OP here, thanks for chiming in. Mostly responding to say that I love your product Loom, use it several times a week.

> If you're taking VC money, you know the ride you're in for. > Don't be naive. And then don't be soft.

Exactly. Eyes wide open, make sure you know what you're choosing.

OP here. My point is that getting paid yearly, especially if a lot of your subscribers pay yearly, means you're back to a similar feast-or-famine kind of cash flow situation as when you're transactional (e.g. an e-commerce shop). This is more hassle since you now need to be more careful about managing your cash flow and making sure you put aside for the leaner months. I fully acknowledge that it is a great hack for customer-funded startups (we do it ourselves at CrankWheel); as with all the "rules" it's more about becoming aware of the hassle associated with a given decision.

I also question whether annual billing increases retention, when your retention is good already (say, 3-5 years projected lifetime). Sure, it lifts your retention when your average lifetime is less than a year, but does it really when you have very low churn? Or are those big annual bills going to be scrutinized more thoroughly than a monthly cost?

I don't have direct experience with Paddle so I don't know whether it's that black and white. Having however experienced a months-long hold on our payment processor account at one point because of an abnormally large charge (a fairly large customer that wanted to pay annually by credit card), I think it could be worth leaving that kind of hassle to somebody else, and have that somebody else be a bigger player that more likely has a better relationship with their payment processors.

There's some operational cost (for running servers and such), but not that high.

There's considerable customer support cost.

The main thing that I feel makes life tough is that the pipeline from, say, doing some paid ads and seeing the results is several months long, and the feedback loop on changes that can affect conversion rates and monetization is similarly very long. We've learned to cope with it but it would be nice to have a one-month feedback loop or shorter.

My company operates on a freemium model. It's a big hassle, as documented in the section about not doing freemium, but it's also something we decided was fundamental to our main distribution channel which is the Chrome Web Store (breaking another of the rules in the article - but maybe check the last section).

I haven't run a non-freemium SaaS company, but I do see what benefits it would bring if we were able to operate on a typical 14 or 30-day trial model. The big ones that would bring are a much shorter sales pipeline and shorter feedback loop on ad spend.