In a sense the simplicity of SLS is a trap: immediately when you need to move past the synchronous lambda invocations via API GW use case (caching, service integrations, step function workflows etc) you need to either fall back to plain CloudFormation or rely on third party plug-ins with possible problems with maintenance, quality and feature-completeness. This makes it a difficult choice to recommend beyond simple use cases.
Yeah this is one area it really struggles. I built a harness (using SST, actually!) for things like step functions, background jobs, cron-ed tasks that I'll try to open source once I clean it up a bunch.
I kind of see that as an unforced error on SLS's side as well; AWS EventBridge is pretty simple and would make those types of workflows possible, but the integration with SLS is really rough.
I loved the folder view! It really highlighted the shift to a modular design in Plasma and solved the problem of desktop eventually ending up as a garbage dump of stuff that at some point was relevant but now impossible to find by providing configurable views to files that exist elsewhere and can be adjusted to fit current needs.
The 4.0 release was otherwise painful though as it was a slow, constantly crashing mess of bugs. At around 4.3 it became stable enough to serve as the primary desktop but IIRC that took years and I saw a lot of friends adopting Gnome or XFCE.
Overload is awesome! However I wonder if some remnants of 90s game design like tight time limits and repeating enemy ambushes that make the game such a familiar and intense experience for the old school in reality disadvantage the game and the genre from reaching wider audiences.
There was also another classic Descent contender, Forsaken, that got remastared in 2018 to run on Linux and macOS in addition to modern Windows platforms. The original game was actually used as a graphics benchmark for early 3d accelerators due to its lighting effects.
That said, looking forward to playing Descent 3 on a modern platform!
Making it a 6DOF shooter at all probably severely hampered any chance of achieving mass appeal. If it had been 40-80 hour open world, character-based, third-person action adventure game, their chances would likely have been much higher.
Yes, I was just rather facetiously making a point that the game is all the more interesting to its intended audience for not trying too hard to appeal to anyone else.
The only Windows box I have is an AMD-based gaming rig and last time I checked the hardware isn't Windows 11 compatible. Time to check out how the gaming experience on Linux is these days.
It's good enough I haven't even started my gaming rig for a few years now... Not painless, but my biggest problems were the System Shock remake that couldn't start a game (an update automagically fixed it) and terrible performances under Lutris with Planet Crafter (but somehow running it within Steam solve the problem). I also do not play the latest games though.
Nearly every game will just work, unless you play a Competitive Game with a kernel anti-cheat. (Looking at you valorant). Https://protondb.com search for the game(s) you play, and hey presto it's fine.
Obviously it's primarily targeted at steam games, Lutris will sort you out with any other windows game (a game manager like steam but for _nearly every_ game)
Not the best way to look at it. A lot of developers today work primarily on github because it's not possible to interact/work in other projects from their siloed forges. When gitlab implements AP, things are no longer siloed and therefore there will be a percentage of (current) Github users who will free to choose their forge independently of network effects.
It's better to ask "how many projects are stuck to Github exclusively because of collaboration tools, and how many of these would be able to leave Github once more Gitlab/Gitea/Forgejo are able to interoperate? How much of its customer base can Github afford to lose before being forced to give up their monopolistic strategy?"
It's approximately 0. Microsoft knew that GitHub had no moat being what it was when they acquired it.
T hat's why they've been adding CI, Pages, Codespaces, Copilot, and a bunch of other crap that only some of which people actually asked for in order to raise switching costs. That would go against everything they've been doing the last couple years.
I'm pretty sure that's only if you're a gatekeeper, which are only the services big enough that my mom would notice directly if they'd go down (things used by millions of people in my country of 18 million). Whether Apple's messaging system, default enabled on every Apple phone, qualifies is currently being debated, to give a sense of what scale this requires.
The DMA/DSA laws also contain rules for smaller parties (I've been getting tons of ToS update emails mentioning the digital somethings act), but not interoperability
- ld-signatures is now W3C vc-data-integrity:
"Verifiable Credential Data Integrity 1.0
Securing the Integrity of Verifiable Credential Data" https://www.w3.org/TR/vc-data-integrity/
- vc-data-integrity specifies how to normalize the document by sorting keys ~ in the JSON before cryptographically signing the transformed, isomorphic graph
- SLSA.dev also specifies signed provenance metadata (optionally with sigstore.dev for centralized release artifact hashes), but not (yet?) with Linked Data
> Blockcerts is an open standard for building apps that issue and verify blockchain-based official records. These may include certificates for civic records, academic credentials, professional licenses, workforce development, and more.
> Blockcerts consists of open-source libraries, tools, and mobile apps enabling a decentralized, standards-based, recipient-centric ecosystem, enabling trustless verification through blockchain technologies.
> Blockcerts uses and encourages consolidation on open standards. Blockcerts is committed to self-sovereign identity of all participants, and enabling recipient control of their claims through easy-to-use tools such as the certificate wallet (mobile app). Blockcerts is also committed to availability of credentials, without single points of failure.
- [ ] SCH: link a git commit graph (with GPG signatures) with other linked data of an open source software project; for example (SLSA,) build logs and JSON-LD SBOMs.
- >> Is there an ACME-like thing to verify online identity control like Keybase still does?
My current primary workstation is very low spec and old compared to a M2 Mac. On that machine Sublime Text is very performant and provides an awesome code writing experience through LSP. I use it as a daily driver with TypeScript, Go and Dart. I try out IntelliJ (or Android Studio) and VSCode occasionally to check out how they are these days but they feel sluggish in comparison.
If you need more IDE-like features like running tests and builds or managing for example mobile phone emulators from the editor it requires more fiddling with configuration and plugins but is doable.
