Generate code using LLM on a small project, then copy manually toward your big project. This reduces the context (and the cost) for the LLM and gives you many small projects where you can experiment ideas with more agility.
Works perfectly fine with firefox snap ubuntu 152.0.4. I can select mic from one web cam and image from another. Is there somewhere a technical description of how they have used API to implement this ? Open sources ?
Works fine for me. I also took a zoom call from Firefox just yesterday (153.0b13 on mac). So, I think this is false alarm.
I also see no good reason why Zoom would want to restrict this. Likewise, I don't see them dropping support for Safari either (works fine as well).
Maybe they limit things on mobile? But on IOS they'd be stuck having to support the Safari browser engine anyway as neither the Chrome nor Firefox browser engines are used on that platform (both browser delegate to Apple's browser engine there).
It is an Azure (and WSL) specific Linux based on a general-purpose Linux (fedora). Having this general-purpose foundation will give access to many packages.
My first job in 1994 was using RDB, very soon updated to RDB-Oracle 6 that was very compliant with my book about sql 92. I learned SQL during my studies, so my queries based on joins naturally performed well, unlike the nested SELECT statements in the legacy code. I learned to tune performances. After that, I've encountered Oracle with its share of non-standard syntax and behavior. Now, when I build a backend (flask or fastapi), my CLAUDE.md contains "use sqlite without sqlalchemy" so that I can understand how data is accessed and check that there is no hidden query inside a loop. When the SQL queries generated by AI become unreadable, it’s a sign that the system is no longer functioning properly. In my view, SQL is a good way to maintain control over an application that is largely generated by AI.
I think the current intellectual property system is flawed. Books are knowledge, and we shouldn't be able to limit the spread of knowledge. I imagine that books could be sold at the cost of printing, and there could be a QR code inside so that readers could freely donate money to the author if they enjoyed the book. Strangely enough, I imagine that with such a system, authors would be better paid.
I think power users are not the main target of Ubuntu.
I have put my parents on Ubuntu (gnome) in 2013 to replace windows XP. My mother is 88 now. I think it is the perfect fit for her (dad is dead years ago).
I use ubuntu gnome because tweaking my computer is not where I want to spend my time. YOLO. Using a "mainstream" desktop that can be explained to "non specialist" has its benefits. I accept to suffer some annoyances and there is always a way to fix the most annoying ones by sacrificing time.
>I think power users are not the main target of Ubuntu.
Then who is? Normies buy iPads and casuals stay on Windows. Is this why Linux can't gain any market share?
IMHO, Ubuntu is trying to gain market share by targeting non-experts — making Linux simple enough for normies and casual users. Casual users are generally less likely to mess things up on Ubuntu than on Windows.
Every time I run Ubuntu on a computer it always ends up in a state where it does not boot after a few months. This has happened on multiple computers, none with nVidia GPUs, over a period of a bit over 15 years. I don't do anything funny with my computers. No custom kernel, no weird kernel modules, no trying to shoehorn in 3rd party repos intended for Debian, etc. The last time I tried was last year, when I got a new job and my work laptop came with Ubuntu 24.04. Sure enough, after a few months an update made it unable to boot. I have not had this problem with any other distro. I switched the laptop to Fedora and it's worked perfectly fine. This makes me question the logic of trying to give Ubuntu to novice computer users.
This isn't true any more, and hasn't been for some years, you know.
It was true but times change.
Microsoft chose to kill off Windows 10, which it once promised would be the last desktop Windows ever. Its replacement is bigger, slower, stuffed with adverts and upselling attempts, and has an artificial demand for TPM 2.
That's driven thousands of people to check out Linux, and if you don't know anything about Linux, then Ubuntu is the number one best-known distro. Many techies dislike Snap (to the extent of spreading lies like "it's not FOSS"), but it makes version upgrades safer, which matters more to non-techies.
(I say thousands so the pedants don't shout at me, but I suspect the reality is at least hundreds of thousands, maybe millions.)
Linux Mint is friendlier, yes, and so is Zorin OS, but both are based on Ubuntu.
Valve has sold millions of Steam Decks, which demonstrate that it's now possible to run premier new Windows games on Linux with performance at least as good as on Windows. All Linux users know their hardware runs faster and cooler with Linux than Windows anyway.
Chromebooks (which are as cheap as laptops get) outsold Macs (which are expensive) by revenue in 2017 in the USA and within 3 years in the rest of the world. ChromeOS is a desktop Linux, based on Gentoo. It has hundreds of millions of users who have never heard the word "Linux".
Companies with cloud-based IT are deploying ChromeOS Flex as a response to ransomware attach. (E.g. Nordic Choice hotels.)
Many of us see Ubuntu's characteristic desktop in shops, bars, travel stations and things regularly now. I hear its startup sound on trains. I have totally non-techie friends running Ubuntu at home. I've given Mint to lots of mildly technophobic friends and they get on just fine.
It's not over, but the year of Linux on the Desktop came about a decade ago, and the penguin taleban were too busy in-fighting to notice.
>That's driven thousands of people to check out Linux, and if you don't know anything about Linux, then Ubuntu is the number one best-known distro.
Not really. Most normies seem to choose meme gaming distros based off Fedora like CachyOS or Bazzite. Many are waiting for official SteamOS release which is based on Arch.
Snap is TERRIBLE for non-technical people. Imagine installing an image editor via Snap, and then the default sandboxing making it unable to access the images on your media drive. No errors, it just silently fails.
This has been a problem I’ve dealt with on nearly every single Snap I’ve installed. If you’re a file editor, you must let me edit my damn files!
I often hear things like this, but I never encounter them myself.
I've run every single version of Ubuntu ever released. Work machines stay on LTSes, testbeds run interim versions.
After the 22.04 release, I carefully de-snapped my work laptop, using `deb-get` to install native packages of everything. Worked a treat, took less disk space, things started a tiny bit faster.
Then I enabled Ubuntu Pro and it force-reinstalled snapd. It's fair enough to have it as a dependency: it's a standard component. I was very annoyed, though.
But when I upgraded to 24.04, a lot of things broke. I had to spend ages re-enabling repositories, getting new keys, changing version strings in stuff under `/etc/apt/sources.list.d` and so on. It's a PITA.
So I have performed a volte face. I removed all my `deb-get` packages, and reinstalled the snap versions. All my comms and messaging apps, music and media players, and so on.
It's much easier. No extra repos. I experimentally took one laptop from 24.04 to 24.10 to 25.04 to 25.10 and yesterday to 26.04. All my apps stay in place. Nothing broke. No custom repos. No changes needed to any config file. It just works.
I've been using Linux for 30 years, starting on Slackware and moving to Red Hat and Caldera and SUSE via lots of others. But I'm old and grumpy and I want stuff to work without fiddling. I want low maintenance. Snap is low maintenance. My messaging apps can download stuff into my Downloads folder, open attachments from Documents, and so on.
I run native packages of my own browsers (Waterfox and Chrome) and AppImages of Panwriter and Logseq, and I have none of these difficulties.
Life is easier if you don't fight the OS and the vendor.
And Ubuntu is still easier and less hassle than Debian, Fedora, openSUSE, Arch, or any of the other big names.
I was student between 1990 and 1993 and Ada was the main language. Compilation speed was not an issue. I remember that Eiffel was very slow to compile, but not Ada. Between 1994 and 1999, I have worked with Ada on Vax machines. The full recompilation took 2 hours because the machine was slow, not because of the language. Other languages were similarly slow (pascal, C). C was slow because of the lack of precompiled headers (many headers had to be parsed many times). With Ada (alsys ada), there were "libraries" that were black boxes directories containing object code and already parsed package specifications.
Between 1999 and 2002, I have handled projects in Ada, C++ and Java. C++ was slightly slower than Ada (slow link). Java was a lots faster.
Nowadays, Ada compilation is faster than C++.
also "Make America Great Again" states that America is not currently great, which given its geo-political and economic position is just dishonest. Combined with "America First" you get an entirely clean canvas to be incredibly radical while cosplaying conservative.
At that time, there was almost no spam because we could report them to abuse@domain...
There was no firewall in front of our campus and we were using rlogin to connect outside. I used export DISPLAY from Brest to Paris (to use xdvi that was not installed locally).
IMO, the security (ssh and killing rlogin) is the main change that is a really useful progress.
