DistroWatch Weekly |
DistroWatch Weekly, Issue 1053, 15 January 2024 |
Welcome to this year's 3rd issue of DistroWatch Weekly!
The concept of voice powered, helpful computers has been in the public eye for decades. Various science fiction books and television shows have portrayed talking computers which could understand speech for longer than society has had personal computing devices. These days talking smart phones, voice-directed speakers, and laptops that can perform searches from spoken words are becoming increasingly common. While there are many commercial devices and proprietary applications which can interpret and respond to human speech, there hasn't been as much progress in this field in the open source community. This week we begin with a look at AI assistants and voice-controlled assistants in the open source realm. Read on for some first looks at these so-called smart assistants for Linux users. Do you use a voice assistant or AI assistant on your devices, one which is either open or proprietary? Let us know in this week's Opinion Poll. Then, in our News section, we talk about new developments happening in the UBports community and report on Qubes certifying a new laptop for the Qubes OS operating system. Plus we share views from a developer of Adelie Linux which explore de facto versus traditional standards. Plus we are pleased to share the releases of the past week and list the torrents we are seeding. We wish you all a wonderful week and happy reading!
This week's DistroWatch Weekly is presented by TUXEDO Computers.
Content:
|
Feature Story (By Jesse Smith) |
Linux AI voice assistants and talking to myself
Up until December 2023 I'd never used a voice-controlled AI assistant. Certainly, I'd watched (or heard) them being used. For years I've observed people talking to Siri on their iPhones or to the Google assistant on Android devices. More recently I've encountered people calling out instructions to Amazon-powered Alexa speakers. However, I'd never decided to talk to one myself.
I suppose I was late to the party in part because I tend to be a quiet person by nature, or maybe it was just because I usually don't add gadgets to my home. Perhaps I've been a bit reluctant to invite a device, whose job is literally to listen in on what I'm saying and send that information to a large corporation, into my home. Probably a combination of all of the above.
Also, and this might just be showing how long I've been using computers, but to my mind it's easy (and less disruptive to the office) if I tap out the following series of key strokes rather than calling out requires to an AI assistant to scheduling a meeting:
F12
at 14:00
notify-send -t 10000 "Go to the meeting"
To me, the above seems more efficient than saying out loud, "Hey, Google. Remind me at two o'clock to attend the meeting." But maybe that's just me.
Anyway, around the holidays I was invited to try out a new Alexa speaker and the experience, well, it didn't go smoothly. My "conversation" with the speaker went approximately like this:
Me: Alexa?
Speaker: I am here. I will start listening when you say one of my key names.
Me: What are your key names?
Speaker: I will respond to Alexa, Amazon, Computer, Echo, or Zippy.
Me: Computer... Computer?.... Echo? ... Zippy? ... Alexa?
Speaker: (beep)
Me: Alexa, what is the significance of the name Zippy?
Speaker: Zippy is defined as short for zipper, a tool for adjusting clothing.
Me: Alexa. I would like you to respond to the name "Computer".
Speaker: I sorry, I don't understand.
Me: Alexa. What features do you offer?
Speaker: I don't understand.
Me: Alexa. What functions can you perform?
Speaker: I don't understand the question.
Me: Alexa, can you play music?
Speaker: I have access to a wide range of music. I've even written some of my own songs, including a rap. Would you like to hear it?
Me: Yes.
Speaker: ...
Me: Alexa, I would like to hear you rap.
Speaker: I'm sorry, I do not understand.
Me: Alexa, I would like to hear the music you wrote.
Speaker: Playing your favourites playlist from Spotify. (Music begins.)
Me: I don't have a Spotify account. Where is this playlist coming from? Alexa, stop. Alexa, what is the weather outside?
Speaker: According to (insert source here) the temperature will be between 32 and 45 degrees Fahrenheit today.
Me: Alexa, what is that in Celsius?
Speaker: I do not know what the weather will be like in the town of Celsius.
Me: You've got to be kidding me. Alexa, what is the temperature here today in Celsius.
Speaker: ...
Me: Alexa, what historical events happened on this date in history?
Speaker: On this date, December 7th, 1941, the Japanese attacked Pearl Harbour.
Me: Alexa, today is December 8th.
Speaker: Today is December 6th.
Me: Alexa, today is December 8th, what historical events happened on December 8th?
Speaker: A flight from Newfoundland crashed, killing all on board.
Me: This is as depressing as it is frustrating.
During my conversation with the speaker I also discovered something interesting: the Alexa speaker will provide statements which it should know are false. I don't just mean it will recite information from a third-party website which is incorrect, but it will appear to check information locally which it then gets incorrect. For example, the following back-and-forth took approximately one minute and revealed a few inaccurate responses:
Me: Alexa, what was the last question asked of you?
Speaker: I have not heard anything for the past minute. However, you can check all past queries under the Privacy section of your app settings. (This was true.)
Me: Alexa, what was the last question put to you?
Speaker: I have not heard anything for the past minute. However, you can check all past queries under the Privacy section of your app settings. (This was false as it was the second time I'd asked in under 20 seconds.)
Me: Alexa, what does the blue light on your speaker indicate?
Speaker: From (source) - A blue light indicates pairing with a Bluetooth device is enabled.
Me: Alexa, what was the last question you were asked?
Speaker: The last query was, "What was the last question put to you?" (This was also clearly false since it ignored the query about the blue light, but the earlier request was recent enough to still be recorded in its memory.)
As you might imagine, I was not sold on the usefulness of a voice-powered assistant, at least not one from Amazon. However, it did get me wondering about open source and Linux-powered assistants and how they might compare against commercial offerings. I decided to go on a hunt through the, to me, largely unexplored digital AI assistant territory and see what I could find that would run on my Linux-powered computers or maybe my PinePhone running UBports. I invite you to come along for the trek.
* * * * *
Linux Assistant
My first instinct was to look up portable, pre-packaged AI voice assistants. I headed over to Flathub and performed some searches. One of the first items to come up was called Linux Assistant. Not much information on this package was offered, other than it appears to be a desktop package that offers a number of graphical wizards to perform common tasks. There doesn't seem to be any language processing from this application, either from the keyboard or through voice commands. It just has a search feature and a series of wizards to simplify system administration.
* * * * *
Almond
The next Flatpak bundle I found was for a project called Almond. Almond appears to be the creation of a group at Stanford University, however I wasn't able to get any details on this application as its website was off-line and it seems the project has been discontinued.
* * * * *
Mycroft
I didn't find any other promising-looking applications on Flathub and turned to more general web searches. One of the first projects to catch my attention was Mycroft. Mycroft stood out because I have heard about this voice-powered assistant before and I've heard it is a strong, up and coming open source project with Siri- and Alexa-style capabilities. According to the Mycroft website the service can handle voice and text processing and requires devices to be linked to an on-line account on the Mycroft servers. Accounts and basic functionality are free, but some add-ons can be purchased to extend the AI's skillset. The Mycroft project also sells smart speakers which can act as dedicated appliances that link to Mycroft accounts.
There don't appear to be any traditional or portable packages for Mycroft (I couldn't find any Deb, RPM, AppImage, or Flatpak packages. There are references to a Snap package, but according to the reports I found on the forums, it's unmaintained.) I did find setup instructions for Linux users. This basically involves downloading the project's source code and running a script which will automate fetching dependencies and set up the Mycroft client service. We can then create an on-line Mycroft account through the website and link our computer to the on-line portal.
I ran the setup script. It began by asking if I wanted to use the project's development branch or the stable branch. I was also asked if I'd like to enable automatic updates. I chose the stable branch with automated updates. Next I was asked if I'd like to build and run audio processing software (called Mimic) locally or rely on a remote server to handle audio processing. I opted to keep my build light and less complicated by relying on the remote server. The script then offers to add the Mycroft executable programs to our PATH variable, which I accepted. The script then fetches some dependencies and starts building Mycroft.
The installation ran for a few minutes and eventually failed with a long Python backtrace. While I was unable to find any pre-built packages for Mycroft, I did find instructions for setting up a Docker container which would allow me to run the client locally. I installed Docker and followed the instructions. While the container was successfully fetched, it failed to run and did not produce any log file. I suspect the container may be unmaintained, based on the issue reports I found while browsing for pre-built package and setup help. Requests for portable packages and reports of having trouble getting Mycroft running seemed to be common, and it seems to be the weak point in this software project. I suspect the developers are focusing on their dedicated speaker appliances rather than unofficial third-party clients. This brought my attempts to run Mycroft on my own to a halt.
I did reach out to the Mycroft team through their on-line contact form when I started this overview of voice assistants and asked about getting a demo model of the Mycroft smart speaker for a trial run. After two weeks I still have not received a response.
* * * * *
Jarvis
The next project which came up in my web searches was called Jarvis. Jarvis is described as a voice assistant which can be used to open websites, adjust media volume, look up information on-line, set alarms, spell words, and check the weather.
The install instructions for Jarvis require users to pull the software from GitHub and then run a setup script, not unlike the Mycroft approach. The setup script fetches some Python packages. In fact, it pauses several times to ask permission to install new packages from Debian repositories (I was running Debian 12 for the purposes of this experiment.) Once the Debian packages had been placed on the system, the Jarvis script automatically begins to fetch more Python modules through Pip. This process eventually failed due to a missing dependency called nltk. This brought yet another experiment to an early close.
* * * * *
Jarvis (again)
I then moved on to another project, this one also written in Python and also called Jarvis. At first I thought this was a clone of the same project, but the install process was a bit different and the output from the setup and client software was different.
This version of Jarvis claims to offer many functions and dozens of recognized commands. The project claims commands and responses can be offered through both text and voice. Unlike the previous projects, this Jarvis installed cleanly and did so without requiring root account access (which the earlier Jarvis and Mycroft had needed).
The second Jarvis can be run from the command line and prompts us with a welcome message which tells us we can run "help" at any time to see a list of commands Jarvis will recognize. We're also told we can type "enable sound" to turn on voice output and speech processing. The text "Hi, what can I do for you?" then appears on the screen. So far, so good.
I'd like to spend a little time talking about Jarvis II's capabilities, starting with its command line interface. Jarvis uses a fairly simple command line syntax, similar to a Linux shell. We can't type in an English sentence and get back a response, we need to provide a key word and, usually, a parameter. Typing "help" provides a list of key word commands. Typing "help" followed by a key word will indicate which parameters the command accepts.
There are several dozen commands Jarvis will recognize. Some of these are simple fun. For example, we can type "joke" to see a joke printed on the screen. We can type "roll" to see a random dice roll in the range of 1-6. There are other, more complex and useful commands. For example, there is a command called "solve" which will try to interpret a mathematical equation and solve for x. For instance, "solve 5 + x = 8" will reply with "3".
There is a command called "remind" which will pop-up a desktop notification at a specified time. One command called "ip" will look up local and remote IP addresses for the current machine. The "movie" command will help us look up information about a given movie. For example, running "movie year die hard" replied with "1988", the year in which the film was released.
Some commands were more of a mixed experience or didn't return the expected results. For instance, the "weather" command successfully provided the current temperature and sky conditions in my area. However, the weather forecast command failed and caused a Python traceback. The "stock" command for looking up stock market info similarly failed. Trying to keep up with current events with the "news" command outputted an error saying Jarvis was missing an API key for this feature.
The "lyrics" command to look up song lyrics failed to find any of the song titles and artists I requested. I couldn't get Jarvis to give me a "quote of the day" either. One command, called "mood", confused me. The help text says we can run a command line "mood music", though it doesn't say what that accomplishes. When I tried to run "mood" on its own or "mood music" the command failed.
All of this is to say, Jarvis has some good capabilities and it has some very basic command processing through its text interface. However, not all of the skills are working and the help text often lacks examples and explanations.
The Jarvis project reportedly works over voice interfaces, receiving and outputting audio. The voice processing is turned off by default, but we can run a command to enable audio processing. The audio output from Jarvis has two modes. By default, it will use a built-in text-to-speech voice which is processed locally. This works, but sounds very stiff and robotic. We can also type a command to Jarvis which will switch to using a Google-supplied voice so that Jarvis sounds like the "OK, Google" voice assistant. However, I found that sometimes Jarvis would crash when the Google voice processing was enabled. The local voice processing seems more stable.
The text output feature reads the text which Jarvis displays on the screen. This allows us to type commands and hear the response, either in the locally processed mechanical voice or the more natural sounding Google voice.
When it came to issuing instructions to Jarvis over voice commands, I didn't have any luck there. I made sure audio processing was enabled, then confirmed my microphone was working and picked up by other applications. However, I was unable to get Jarvis to indicate it was hearing me. I tried speaking the commands as I'd type them as well as speaking the examples shown in the Jarvis documentation. However, after five minutes I had to acknowledge I was just talking out loud to myself.
* * * * *
Jarvis (yet again)
During my week I went looking for additional assistants and discovered at one more called Jarvis. This one claims to be an AI text assistant and a chat bot. This third Jarvis reportedly runs in web browsers, on Android devices, and on Windows. It wasn't the style of assistant I was looking for, so I didn't download it.
* * * * *
Brave's Leo
Finally, I'd like to acknowledge the Brave web browser has recently launched a version with an AI assistant built into it. I installed Brave and tried out the assistant, called Leo.
The Leo bot can be accessed through the browser's side panel. It seems to be aware of the content of whatever browser tab currently has focus. However, the AI doesn't have access to the Internet outside the current tab and doesn't have features which can tie into functions of the underlying operating system. In other words, Leo can converse with us over text, "discuss" some topics, and try to summarize web pages for us, but it can't work through voice commands, play music, look up information, set timers, or perform other AI assistant style functions.
While the Leo bot provided well written responses, it mostly answered my questions with some variation of "I'm an AI assistant without access to the Internet and cannot provide this information or cannot provide an opinion on this subject."
* * * * *
Conclusions
What I discovered this week from diving into the sea of AI voice assistants for Linux is that, for the most part, there aren't any good, accessible, voice-capable assistants yet. Though the issues behind my evaluation vary. Mycroft certainly looks like it is highly capable, complete with a dedicated speaker system and a friendly web interface for extending capabilities. However, even on a supported distribution like Debian, the install process fails. Further, Mycroft (like most of the projects I've just talked about) doesn't offer a native or portable package. It requires technical knowledge to set up and, even then, doesn't necessarily succeed in setting itself up.
This seemed to be the most common thread in my experiments: the AI voice assistants - whether they had good capabilities or not - almost all suffered from a lack of up to date, useful documentation and usable packages, even on distributions they reportedly support. This means there is a high barrier to people trying out these assistants.
The ones which did install, like Jarvis and Brave, are interesting, but still in their early stages. They have a few useful features and I'm eager to see how these technologies evolve, but they still feel crude and a bit awkward to use compared to dedicated applications or widgets with a narrow focus. The AI revolution may be coming, but it has not arrived on desktop Linux yet. On the other hand, I did get more useful information out of Jarvis than I did from Alexa - a technology backed by one of the world's wealthiest corporations - so I think there is some hope on the horizon for voice assistants on Linux.
|
Miscellaneous News (by Jesse Smith) |
UBports provides overview of coming changes, Qubes certifies StarBook laptops, Adelie developer discusses written versus de facto standards, Asahi Linux improves energy savings
The UBports team has published a news and Q&A post which covers some of the changes the developers expect to see in the coming year, along with responses to questions from the community. There is also a summary of some of the little changes happening behind the scenes: "In news, Cosmia is trying to port Lomiri to NixOS. There have been some additions to Teleports from Florian and Dan. LionelD has been doing work on import/export. Rachanan is in the process of upgrading qtwebengine. Some security updates have been put in place. Mohammad is working on a dark mode switch for the System. He is also adding a configurable charging information alert to the lockscreen. Rather than just % it will be able to say '3 hours 40 minutes until charged'. LionelD is working on notifications which are shown but blurred out for privacy."
* * * * *
The Qubes project has certified the Star Labs StarBook as being compatible with Qubes OS 4.x. "It is our pleasure to announce that the Star Labs StarBook is officially certified for Qubes OS Release 4! The Star Labs StarBook is a 14-inch laptop featuring open-source coreboot and EDK II firmware. In addition, the StarBook is currently the only Qubes-certified computer with out-of-the-box support for qubes-fwupdmgr, a new feature in Qubes OS 4.2 that allows Qubes OS to securely update the computer's firmware." Certified devices are not only compatible with Qubes OS, but can be purchased with Qubes pre-installed. Details on the StarBook and the Qubes certification program can be found in the Qubes OS announcement.
* * * * *
The Adelie Linux team develop a desktop and server distribution which strives for clean design and standards compliance. One of the developers has weighed in on some of the challenges which face distributions that stick to standard-compliant libraries and tools (such as the musl C library) versus distributions which use de facto standard packages like glibc. The focus of the post is on systemd, both the benefits it brings and some of the complexity it adds to a distribution: "I am also unhappy about systemd's lack of musl libc support. That is probably a blessing for me, because it's an easy reason to avoid trying to ship it in Adélie. While I have just spent five paragraphs noting how great systemd is at service management, it is really bad at a lot of other things. This is where most articles go off the deep end, but I want to provide some constructive criticism on some of the issues I've personally faced and felt while using systemd-based machines.
* * * * *
The Asahi Linux project is working with Fedora to develop a Linux distribution that runs on Apple M-series computers. The project has made progress on many fronts, improving the day-to-day usability of the distribution. One area where the project has made progress is with energy savings: "While working on speaker support, we found that Pipewire and Wireplumber were constantly being mis-scheduled onto P-cores. By default, the kernel prioritizes never being 'late' above everything else for real-time threads, thus audio processing, due to its real-time nature, was always being given full performance. We did the math, and we found we don't need anywhere near full performance to run our DSP code. To fix this, we gave Pipewire and Wireplumber the ability to use utilisation clamping, a scheduler feature that lets applications peg their performance requirements to a fixed range. We cap Pipewire and Wireplumber to an extremely low maximum performance so the scheduler restricts them to efficiency cores at their lowest operating point. Both still function perfectly, and we get to save oodles of battery life for our users! This awesome feature goes so underutilised that it wasn't even enabled in the standard Fedora kernel until we asked for it to be a couple of weeks ago (CONFIG_UCLAMP_TASK), and we sincerely hope that its enablement in Fedora leads to more widespread adoption." Additional updates are provided in the Asahi Linux blog post.
* * * * *
These and other news stories can be found on our Headlines page.
|
Questions and Answers (by Jesse Smith) |
Do some distributions run hotter than others?
Some-like-it-hot asks: Have you ever encountered a distribution that runs hotter than others? For instance, I recently installed MX Linux on my primary computer and observed that the CPU fan would continuously fluctuate in speed, ramping up and down every 5-10 seconds, even when the system was idle and no applications were running. Upon executing the sensors command, it was confirmed that MX Linux ran hotter than openSUSE Tumbleweed while idle.
I then installed pure Debian and while it did not run as hot as MX, it still ran hotter than openSUSE Tumbleweed. Also, running Debian did not cause the fan to turn on. The load averages for MX and pure Debian were about the same while openSUSE was slightly less.
What would cause this discrepancy and have you ever experienced this before? Do different distributions really run hotter than others? I would be concerned that a distribution would run so hot as to be detrimental to the hardware it's installed on.
DistroWatch answers: Let's take these questions in order.
- Have I experience this before? Yes, I have, to varying degrees. (Pardon the pun.)
- Do different distributions run hotter than others? Yes, they can. There are a number of factors at play. Basically, the computer will get hotter with the more work it needs to do - usually because the CPU is being used more, but the video card can also be a factor. The important thing is: the computer will get hotter when it needs to work harder.
What causes the computer to work harder, especially when it seems to be sitting idle? Usually it boils down to one of three things: the video drivers, the desktop environment, and the services or widgets being run.
I tend to see the heat difference mostly on distributions which run desktop environments that are more flashy. Any 3-D desktop (such as GNOME or Cinnamon) or a desktop using visual effects will require quite a bit more effort to run, even when mostly idle, than 2-D desktop environments with visual effects disabled. This situation is compounded when less efficient video drivers are used. The video card is usually tasked with handling 3-D effects, fancy visuals, and such - assuming support for these features is included in the video driver. Video cards are good at these sorts of things and relatively efficient.
When the video driver (or the video card) doesn't support handling these features, the flashy visual effects and 3-D elements are handled by the CPU, which is less efficient. This will bog down the CPU and generate more heat.
Background services or widgets which are constantly updating will also put more strain on the CPU, even when the desktop isn't being used interactively, and this will again cause more heat to be generated.
Going back to the examples given in the original question, I'd expect MX Linux to run slightly hotter than plain Debian running the same desktop. Despite the fact MX Linux is directly based on Debian, the two projects have some notable differences. MX Linux, if I remember correctly, runs more desktop widgets, enables transparency, and has a Conky status panel constantly updating. Debian uses a vanilla theme, no special effects, and doesn't have any active widgets on the default desktop. The load average will probably be about the same on these two distributions, but MX will tax the CPU more and run hotter.
Basically, a quiet, 2-D desktop with up to date video drivers will probably run cooler than a busy, 3-D desktop running less efficient video drivers.
- I would be concerned that a distribution would run so hot as to be detrimental to the hardware it's installed on. This shouldn't be a problem, unless there is a serious flaw in the computer's hardware. In situations where the distribution is running slightly hot, compared to another operating system, it probably shouldn't run a lot hotter unless there is something quite wrong (like running a 3-D desktop without the proper video driver).
Even in scenarios where an operating system is running unusually hot, the computer's hardware should turn itself off before the temperature gets high enough to risk damaging the equipment.
* * * * *
Additional answers can be found in our Questions and Answers archive.
|
Released Last Week |
Solus 4.5
Solus is an independent Linux distribution featuring the eopkg package manager and the flagship edition runs the Budgie desktop environment. The latest release of Solus, version 4.5, introduces a new desktop spin for Xfce (which is considered to be in beta status) and switches to the Calamares system installer to get the system up and running. "The Solus team is proud to announce the release of Solus 4.5 Resilience. This release brings updated applications and kernels, refreshed software stacks, a new installer, and a new ISO edition featuring the Xfce desktop environment. With Solus 4.5 comes a new installation experience using the Calamares installer. This enables easier installation using filesystems like BTRFS, specifying your own partition layout right in the installer, and it's a major step in migrating away from Python 2, which the old os-installer was written in. A ton of work has gone into the new installation experience, and we are glad to finally release it to everyone!" A complete list of changes and an overview of key package updates can be found in the project's release announcement.
Linux Mint 21.3
The Linux Mint team have published a new version of their Ubuntu-based distribution: Linux Mint 21.3. The new point release includes several new features, including a new type of Cinnamon add-on (or "spice") which adds actions to the file manager. This release also includes support for Secure Boot and introduces experimental support for Wayland sessions in Cinnamon. "Cinnamon add-ons are called 'spices'. Until now Cinnamon supported 4 types of spices: Applets: Little programs which can be placed in your panel(s); Desklets: Same thing, but on your desktop; Extensions: Scripts which can alter the way Cinnamon works; Themes: Look and feel for your Cinnamon desktop. Cinnamon 6.0 features a new type of spices: Actions. Actions, formerly known as 'Nemo actions', are add-ons for your file manager's context menu. Right-click an .iso file and look at the context menu. The 'Make bootable USB stick' and 'Verify' menu items are Nemo actions, provided by the mintstick package. In Cinnamon 6.0, actions are now a type of spice. You can download, enable and rate them like any other spices: applets, desklets, extensions and Cinnamon themes." Additional information is available in the project's release announcement and in the what's new document.
* * * * *
Development, unannounced and minor bug-fix releases
|
Torrent Corner |
Weekly Torrents
The table below provides a list of torrents DistroWatch is currently seeding. If you do not have a bittorrent client capable of handling the linked files, we suggest installing either the Transmission or KTorrent bittorrent clients.
Archives of our previously seeded torrents may be found in our Torrent Archive. We also maintain a Torrents RSS feed for people who wish to have open source torrents delivered to them. To share your own open source torrents of Linux and BSD projects, please visit our Upload Torrents page.
Torrent Corner statistics:
- Total torrents seeded: 2,949
- Total data uploaded: 43.9TB
|
Upcoming Releases and Announcements |
Summary of expected upcoming releases
|
Opinion Poll (by Jesse Smith) |
Using virtual and voice assistants
This week we began with a look at virtual assistants and voice assistants running on Linux. We'd like to hear if you have any of these virtual assistants in your home - perhaps running on a laptop, a smart speaker, or your phone.
You can see the results of our previous poll on paying money for a distribution in our previous edition. All previous poll results can be found in our poll archives.
|
Do you use a virtual/voice assistant?
Yes - an open source one on my phone: | 9 (1%) |
Yes - a proprietary one on my phone: | 108 (7%) |
Yes - an open source one on my laptop: | 6 (0%) |
Yes - a proprietary one on my laptop: | 6 (0%) |
Yes - an open source speaker: | 3 (0%) |
Yes - a proprietary speaker: | 74 (5%) |
Yes - another options/platform: | 8 (1%) |
Yes - a combination of the above: | 49 (3%) |
No - none of the above: | 1252 (83%) |
|
|
Website News |
New distributions added to waiting list
- Adelie Linux. Adélie is an independent, libre operating system based on the Linux kernel and musl runtime library. It runs on multiple CPU architectures and is available in both desktop and server editions.
* * * * *
DistroWatch database summary
* * * * *
This concludes this week's issue of DistroWatch Weekly. The next instalment will be published on Monday, 22 January 2024. Past articles and reviews can be found through our Weekly Archive and Article Search pages. To contact the authors please send e-mail to:
- Jesse Smith (feedback, questions and suggestions: distribution reviews/submissions, questions and answers, tips and tricks)
- Ladislav Bodnar (feedback, questions, donations, comments)
|
|
Tip Jar |
If you've enjoyed this week's issue of DistroWatch Weekly, please consider sending us a tip. (Tips this week: 1, value: US$100) |
|
|
|
bc1qxes3k2wq3uqzr074tkwwjmwfe63z70gwzfu4lx lnurl1dp68gurn8ghj7ampd3kx2ar0veekzar0wd5xjtnrdakj7tnhv4kxctttdehhwm30d3h82unvwqhhxarpw3jkc7tzw4ex6cfexyfua2nr 86fA3qPTeQtNb2k1vLwEQaAp3XxkvvvXt69gSG5LGunXXikK9koPWZaRQgfFPBPWhMgXjPjccy9LA9xRFchPWQAnPvxh5Le paypal.me/distrowatchweekly • patreon.com/distrowatch |
|
Extended Lifecycle Support by TuxCare |
|
Reader Comments • Jump to last comment |
1 • Virtual and voice assistants, or just language model (by Vinfall on 2024-01-15 02:12:38 GMT from Singapore)
It reminds me of a discussion between my friend and me, about whether there is ANY truly voice model with underlying voice input and directly voice output.
The surge of LLM certainly boosted so-called voice assistants, as they are fundamentally keyword trigger, phrase analyzer/balancer/responder, and trained language model in general. I'm unsure about which path would lead to a better way, since there is currently no comparable voice model. And making already established TTS (text to speech) and STT technologies better, and integrate them into LLM seems much simpler to me.
2 • Speech to Text (by stanr on 2024-01-15 02:46:19 GMT from United States)
My only interest currently, would be local speech to text. For me, it would be very useful to blurt out my thoughts quickly. Then, I would more likely edit, organize and push them out to my discussion groups.
3 • Virtual/voice assistant (by fenglengshun on 2024-01-15 02:53:27 GMT from Indonesia)
I tried, for about a day, when I first got a wireless earbuds. And it's bad. If it's already bad above, the experience is worse in lower-end devices.
The idea was that Google Assistant should have been able to response to "Hey, Google," from your earbuds or to your phone, at any time, right? Nope, it takes a few tries, then it needs to spend a few seconds loading the interface, then it needs to hear what I say, and half the time because of my accent and the low-power nature of my devices, it just doesn't interpret it correctly.
In the times when it does or I typed instead of spoken the command, most of the time it doesn't really respond satisfyingly because unsurprisingly it's just a google search with no ability to interpret the full implications of my questions and what the search result says.
Things like ChatGPT responds better, but to me they're more a specific-purpose tool -- if I ask AIs questions that aren't specific enough, then it still get it wrong. At which point I just treat it as an advanced filter and text generator/converter tool -- I'm not going to bother asking even ChatGPT questions that's better answered with a normal search, I'm only going to use it when I'm too lazy to make or adjust some bash scripts I need to do batch operations or something.''
Virtual assistants need to just focus on being a good tool instead of pretending it could be the be-all end-all of every questions or requests you want to ask.
4 • Speech to text (by DaveW on 2024-01-15 03:01:19 GMT from United States)
I use spech-to-text on my Android phone (works quite well) but I don't believe that is what you are calling a virtual/voice assisstant.
5 • Voice Assistants (by Wedge009 on 2024-01-15 04:30:53 GMT from Australia)
Don't worry Jesse, I - and I suspect many other DW readers - have not or would not want to use voice assistants. For starters, I imagine many of us would already be quite adept at typing anyway. And then there's the possible privacy concerns depending on the implementation. I was not aware of there being Linux-based offerings so at least it was an interesting read to go through your findings.
The speech-to-text application mentioned by stanr reminds me of one of my previous employers in the late 2000s, using such software to dictate his notes. I found the number of times he kept saying 'scratch that' (to cancel the last phrase uttered because the program misinterpreted his words) amusing, but clearly he found it useful enough to keep using it even though he was proficient at typing anyway (I think perhaps he was trying to reduce RSI-related pain).
I presume the technology will continue to improve with time to the point it might actually be useful someday beyond just being a gimmick or novelty. But at this point I have no desire or interest in trying voice-operated tools.
6 • Adelie (by dkmillares on 2024-01-15 05:36:12 GMT from Brazil)
Glad to see Adelie Linux in the news and on the waitlist.
7 • MX Linux crazy hot (by mixer on 2024-01-15 07:09:00 GMT from Moldova)
I can confirm that MX Linux is a hooot distro. I tried more than 50 distros on my laptop and MX Linux was for ages (8 years) the only one which heated my pc, and for no reason. The aluminium chassis was always hot on top of the location where CPU is situated. It was in every version for last 5 years, including latest one. Interestingly antiX is ok.
I suspect that MX uses aggressive kernel settings...
Another distro which heats my pc is Ubuntu Mate (both 22.04 and 23.04).... Before that Ubuntu Mate worked as charm.
P.S: this is very frustrating, it is literary not clear what to do, cause system monitor shows no processes which eat CPU, it is very confusing for me.
8 • voice assistant (by Luca on 2024-01-15 07:34:16 GMT from Italy)
I really liked the article about the "AI" voice assistants. It's always the same for me: I see people so happy about their new spying device and then I see them having trouble with them XD
9 • MX linux (by A vd Tweel on 2024-01-15 08:27:30 GMT from The Netherlands)
I've been running MX since 2019, and Debian from 2014 to 2019 on my desk PC. I never noticed any significant rise in temperature. I bought a HP laptop in 2023, and put MX23 on it. The only times the fan came on was when I ran DOSBOX.
10 • Voice assistants and AI (by perstreperous on 2024-01-15 09:45:53 GMT from United Kingdom)
I suspect that many Linux users are Linux users so they can omit voice assistants, AI and the rest of it, especially with Microsoft declaring 2024 "the year of AI-powered Windows PCs" (including a key on keyboards to bring up Copilot, its AI assistant) and "a more personal and intelligent computing future where AI will be seamlessly woven into Windows from the system, to the silicon, to the hardware" ...
11 • Voice Assistants (by DachshundMan on 2024-01-15 10:07:41 GMT from United Kingdom)
I do not use a voice assistant as I have no confidence that my data will be kept confidential and that questions that I give it will not be used to power advertising or to train an AI without my permission. After going to the trouble of using anonymous search, ad blocker, Firewall etc why would I then choose to give my data away. Perhaps a 100% local installation could be OK but then how would the developers make their money ?
12 • @ 10 Voice assistants and AI (by James on 2024-01-15 11:01:03 GMT from United States)
I suspect that many Linux users are Linux users so they can omit voice assistants, AI and the rest of it,
Yes, exactly why I use Linux!
13 • Voice assists (by crayola_eater on 2024-01-15 11:07:04 GMT from United States)
Had to choose 'None of the above' as I didn't see listed the voice assistant that I rely on - the voices in my head :-) Like AI, they are gererally useful, yet do experience confusion and hallucinations.
14 • @7 MX heating laptop (by Chris Whelan on 2024-01-15 11:51:30 GMT from United Kingdom)
@7 Have you reported this on the forum? They are very helpful, and may find a solution for you. Alternatively, the MX Package Installer makes it easy to try other kernels, to see if that fixes things. I can't find other reports of overheating on the forum, so it doesn't seem to be a common problem.
15 • @7 - MX heating laptop (by Uncle Slacky on 2024-01-15 12:16:05 GMT from France)
Ensure that you're running thermald and tlp (tlp is installed by default on MX IIRC), that should keep the temperature down. Also, consider cleaning your fan and exhaust vents and reapplying thermal paste to the CPU, especially if your PC is a few years old.
16 • Voice assistants (by dragonmouth on 2024-01-15 13:16:19 GMT from United States)
The price of convenience is security and privacy.
17 • Eh? Aye! (by Friar Tux on 2024-01-15 14:13:53 GMT from Canada)
I think I've mentioned this here before. I look forward to the day when I can call across the room to my laptop to write a letter to Mom, and the laptop AI presents me with a well written draft to peruse and make minor changes to. This is, of course, IF I ever get used to talking out loud to a computer/laptop/device. I prefer to work silently on my machine, BUT an AI assist would be awesome. I used to have a search engine/launcher on my laptop call Cerebro that was quite good. It stopped working a few years ago, and I haven't found a good replacement since. Cerebro did just about everything I threw at it. It converted metric to Imperial, did calculations, got weather reports, got recipes by ingredient lists, and much more. Anyway, it's gone now, but I do have a bunch of bookmarks to site that do most of the stuff, so, no lose. I DID vote "None Of The Above" as I have tried all the programs/apps that Jesse mentioned in his article and found the same issues as Jesse. Mycroft looked to be the most promising but trying to install it was a mess so I gave up. While I really don't actually care if any of these VAs "call home", I DO prefer something local on my machine. I don't have privacy issues as I don't do anything "questionable" on my machine. (Questionable = anything I wouldn't want my Gramma, or kids, to see.) I still haven't yet, but I'm going to try out MakuluLinux's AI assisted distro. Should be interesting.
18 • Do you use a virtual/voice assistant? (by Geo. on 2024-01-15 14:42:29 GMT from Canada)
Absolutely not. I had no privacy growing up in a small town and working in a public business. I absolutely do not want some device that listen in or spy 24/7/365. I feel naked and exposed ever since I gave up my land line. I look for opportunities to abandon my leash (mobile) whenever I can. AI will only be used for evil and gain by nefarious actors. Jesse, thank you, that story was hilarious. 😀
19 • In Answer to Friar Tux from Canada on 2024-01-15 14:13:53 GMT (by RetiredIT on 2024-01-15 14:52:25 GMT from United States)
"The development of full artificial intelligence could spell the end of the human race." Stephen Hawking (1942-2018), theoretical physicist, cosmologist, and author
20 • Voice assistance, internet powered, regardless of operating system. (by Greg Zeng on 2024-01-15 16:25:52 GMT from Australia)
Results from the Distrowatch survey this week show that about 80% of the readers are not using computers correctly.
The voice assistant in Apple, Windows and Android needs internet access. Some might argue that Android-based systems are not Linux, either open-source or closed-source. Therefore Android and hence Linux do not have voice assistance.
Voice assistance may have no worries about the breach of privacy. "What was the average weather in New York, on Christmas day?" Voice Assistance is also used in real-time, on internet-powered computers, regardless if Google Keyboard (Gboard) is used. Voice search on websites can be used when microphone input is allowed for text input.
Most web browsers can easily and quickly use YouTube's "subtitles/cc" application. This enables most YouTube publications to print the voice code onto the screen display in real time. The raw text capture is usually in English. Voice capture can then be permanently stored for further use.
21 • Artificial intelligence (by Bobbie Sellers on 2024-01-15 17:08:54 GMT from United States)
Above some competent individual wrote: "In Answer to Friar Tux from Canada on 2024-01-15 14:13:53 GMT (by RetiredIT on 2024-01-15 14:52:25 GMT from United States) "The development of full artificial intelligence could spell the end of the human race." Stephen Hawking (1942-2018), theoretical physicist, cosmologist, and author "
No worries. The AI being developed is little more than a search engine. Full artificial intelligence requires consciousness and I doubt very much that anyone aside from philosophical theoreticians is interested in that.
Heck some companies seem determined to stamp out customers consciousness. Thus the tools that enable consciousness suppression. Some individuals seem determined to suppress consciousness personally and use chemicals to that end.
22 • Talking to computers (by Bobbie Sellers on 2024-01-15 17:14:45 GMT from United States)
Have you thought to try Knoppix, I believe the version that is voice controlled is called "Ariadne" but it has been years since I paid attention. The problem is that random noise, radio or TV can interrupt you. I prefer typing to talking and am slow at either.
Favorite feature of posting is the chance to garner fresh Twain quotes.
23 • Knoppix (by Jesse on 2024-01-15 17:28:17 GMT from Canada)
@22: "Have you thought to try Knoppix, I believe the version that is voice controlled is called "Ariadne""
You're thinking of the "Adriane" edition of Knoppix. It is not voice controlled. It has a screen reader built into it that runs by default.
24 • AI (by mf6333 on 2024-01-15 17:29:11 GMT from Poland)
Mycroft can be found in AUR. Sometimes I ask some questions to Aria - AI built in Opera web browser.
25 • Mycroft is kaput (successor projects exist) (by ChanceNCounter on 2024-01-15 19:37:22 GMT from United States)
You haven't received a response from the MycroftAI team because they wound down operations about a year ago. Unfortunately, although they handed over their forum to a successor project, the rest of their web site and attached infra (including their GitHub org) is just sitting there misleading visitors.
You were probably looking for its successor NeonAI, a surviving derivative with a staffed studio behind it. There's also a nonprofit foundation upstream of Neon, currently in the process of incorporating, but we aren't in a position where I'm comfortable triggering a sudden influx of users ;)
Note to passing vigilantes: the former Mycroft staffers whose names turn up in old GitHub issues are not the former Mycroft staffers responsible for, well, any of it. Please don't harass random FOSSies! (You wouldn't think I'd need to add this to a random comment on DistroWatch, but we've had some experiences.)
26 • Voice Assistant (by pc4life on 2024-01-15 19:39:00 GMT from United States)
It wouldn't even occur to me to try a VA on my pc or laptops.
27 • Voice assistants, Copilot Linux (by Mr. Moto on 2024-01-15 19:52:57 GMT from Japan)
My experiences with Google assistant mark it as an annoyance more than a helper, popping up unwanted and not popping up when called for. It's now disabled on my phone.
@10, Copilot is available in the Linux Edge browser. I use Edge sometimes, so might try it out one of these days.
28 • AI and Linux (by Simon on 2024-01-15 21:55:46 GMT from New Zealand)
Just at the moment (though not for long), AI still serves mostly incompetent users who can't get things done quickly themselves, and Linux users tend to be the kinds of competent users who can, so the survey results aren't surprising. It's not just the speed and privacy issues that Jesse mentioned explicitly at the start (typing instructions is often faster than delivering them verbally as a natural language request, and to use one of the current AI speakers is to give up your privacy voluntarily)... it's also the unreliability illustrated in the conversation with Alexa: this is very significant and makes these assistants worthless in many contexts.
It's one thing to ask your AI to play a particular song and it plays something else: you know immediately that it has got things wrong. It's different when it's providing answers to questions when we *don't* already know the answers (e.g. "how much do I need to charge that client?"); here the consequences of even one mistake every hundred questions could be very, very bad. Reliability is essential, and we know when we type an instruction that the result can be trusted: we don't, when we ask an AI assistant, so the current generation of AI assistants have limited use.
Give them a few months though, or years at the most. These technologies are improving at a staggering pace... and we're about to see more and more use of edge AI, in which the training that's been done with these online engines over the last couple of years is coded into portable engines that don't need to access the huge datasets (so that you can operate your phone or whatever using natural language instructions but *without* having to go online). As the language processing gets better and better and the privacy issues are mitigated by edge implementations, more and more of us will be using it. AI is going to start popping up everywhere: in video game characters that respond in natural language when we ask them natural language questions; in image editing plugins that select objects intelligently (e.g. a person in a photo) and allow you to drag them around while the space they used to occupy is filled with believable content, and so on and so on. Linux will probably be a playground for some of the most interesting and creative uses of it, in the sciences at least... for a (probably brief) season, before the Linux kernel becomes as irrelevant as all the other IT that used to be developed by humans rather than by the IT itself.
29 • AI assistant (by Dan on 2024-01-16 01:26:50 GMT from Israel)
I've found a free-as-in-freedom AI assistant for Android, called Dicio. It's made by one of the developers of NewPipe, an amazing app I was glad to discover years ago and still use. Dicio is available on F-Droid. I've never tried Dicio myself but I was surprised it wasn't reviewed and nobody in the comments mentioned it yet.
30 • Leo AI on Brave browser (by Andy Prough on 2024-01-16 05:31:00 GMT from United States)
I just downloaded Brave and turned on the Leo AI assistant thing in the browser sidebar. I asked Leo to write a thankyou letter to my Aunt for a christmas gift, and Leo wrote a lovely and thoughtful thank you note.
There's a button to "Summarize this page", and when I clicked it Leo wrote a shorter thankyou note, but still a nice one.
I asked Leo to summarize a sporting event from today, and it wrote about the score from a game that actually never happened. Complete work of fiction.
I asked Leo to summarize the weather in my city today, and it told me that the temperature was "mostly sunny with a high of 78°F (25°C) and a low of 56°F (13°C)". This is complete fiction - the weather was below freezing all day today and we had snow.
I asked Leo to write a python script to print the current weather in my city, and it wrote one up very quickly. I doubt that it actually works, but it looks cute.
I asked Leo how many gold records Taylor Swift has, and it gave me a completely fictional answer that a Grammy Award = 1 gold record, and that she therefore has 10 of them.
I asked Leo what the conversion rate from dollars to Euros is today, and it told me 1 dollar = 0.89 euros. I did a Brave search for dollars to Euro conversion, and Brave search told me 1 dollar = 0.92 euros. So Leo does not have access to Brave's own search engine, and it just makes up completely fictional responses and passes them off as fact.
Leo certainly does not appear to be capable of doing any real work of any kind whatsoever - it appears to be just grabbing some intelligent sounding text from the internet and copying and pasting it and modifying it and calling it artificial intelligence. It writes fictional "news" that never happened and reports it as fact. Nothing that it says could in any way be trusted. If you used this for a homework assignment you would almost certainly get a failing grade.
31 • Any old algorithm presented as "AI", and open voice recognition (by perstreperous on 2024-01-16 09:17:36 GMT from United Kingdom)
@30: Also observed here. This sort of thing is (going to be) a major problem - "generative AI" which is an existing (non-AI) algorithm with a coat of paint.
@29: I am the worst nightmare of English voice recognition, as I am Scottish, but the voice recognition by Dicio is spectacularly (and hilariously) off - annoyingly, it quite often gets words right then has a think and changes them to something else. It reminds me of where Google was about eight years ago. It is unusable here, which is a shame as open, offline tools of this type are desperately needed.
32 • Voice Assistant (by Knightron on 2024-01-16 09:29:12 GMT from Australia)
I feel like the poll isn't really providing enough options considering how common smart phones are these days. I have a proprietary voice assistant on my phone but I personally never use it. I have no interest in voice assistants.
33 • @17 Friar Tuck: (by dragonmouth on 2024-01-16 12:04:48 GMT from United States)
One of the biggest fallacies in computer universe is "I have nothing to hide".
34 • @21 Bobby Sellers: (by dragonmouth on 2024-01-16 12:28:51 GMT from United States)
I'm glad you mentioned "search engines". The first few were developed to provide answers to users. Over time, they became data-harvesting applications. Computers were a quantum leap in data collection and correlation over manual methods. AI will be as big an advance in data gathering and manipulating over computers as commercial space flight will be over hot air balloons.
35 • AI Assistant (by Otis on 2024-01-16 13:08:25 GMT from United States)
The "hey Siri" thing made me laugh the first time I used it, and the response was a list of possible answers I already knew, so I googled it and zeroed in on the surgical technology class of instrument I was looking for.
No more artificial "intelligence" after that. Have fun with it, people. I don't use a microwave oven and both my car and motorcycle are stick shifts. Lemme alone. Scram, ya bother me.
36 • Nothing to hide (by Friar Tux on 2024-01-16 13:13:11 GMT from Canada)
@33 (dragonmouth) if you're referring to me, I didn't say I have nothing to hide. I said I do not do anything "questionable" on my machine. (Questionable = anything I wouldn't want my Gramma, or kids, to see.) I have plenty of stuff I consider personal/hideable/no one else's business, but it is not on my computer. If you're referring to the general population, I might be inclined to agree.
37 • @17; @19; @21--artificial intelligence; quote by Hawking. (by R. Cain on 2024-01-16 17:53:27 GMT from United States)
"...The development of full artificial intelligence could spell the end of the human race..." Stephen Hawking, interview with the BBC, 2014
-------------------------------------------------------------------------------------------------------------
"OUTSIDE OF THEIR PARTICULAR AREAS OF EXPERTISE scientists are just as dumb as the next person."---Richard P. Feynman
"I have found that the reason a lot of people are interested in artificial intelligence is for exactly the same reason a lot of people are interested in artificial limbs: they are missing one."...and...
"Artificial intelligence has the same relation to intelligence as artificial flowers have to flowers." David L. Parnas
38 • hotter distribution (by Matt on 2024-01-16 18:40:18 GMT from United States)
I never tried to find out the cause, but Void linux runs much cooler than Debian on the same laptop. Both were running XFCE and the same proprietary Nvidia driver.
39 • @36 Privacy, not secrecy (by Douglas Reynholm on 2024-01-16 19:00:46 GMT from Denmark)
Even if people have no secrets they still have (or should have) the right to privacy. I draw the curtains at night, not because I'm doing something dodgy (I swear), but because I want to do the dishes, read books etc. in private.
@Jesse: Brilliant and hilarious feature story this week. thumbsup.jpg
40 • running hot (by MikeOh Shark on 2024-01-17 01:27:54 GMT from United States)
I started running MX Linux Xfce version about 5 years ago when I got a new laptop. It ran hot when new but each kernel update dropped my power consumption ( I had my laptop plugged into a wattmeter).
After I installed tlp and powertop I found that my power consumption was much lower. Now, my laptop runs cool and I never hear the fan come on. I use about 5 to 9 watts when surfing with the upper end being when screen brightness is turned up and the keyboard backlight is on. Even when running HandBrake it doesn't get hot.
Ask for help in their excellent forum. Be sure to post your system information as someone will likely have a similar machine and setup.
41 • AI Assistant (by A on 2024-01-17 02:51:46 GMT from Poland)
I'm finding the whole idea of talking to a piece of metal degrading and dehumanizing. We're already living in a dystopian world and it's only going to get worse...
42 • AI (by zephyr on 2024-01-17 10:31:36 GMT from United States)
@37 Back in the late 1960's to early 1970's had a chance encounter with Stephen Hawking. He was sitting rather uncomfortably in a wheel chair and spoke with a voice box, His caregiver translated he wanted to look at a carbon arc projector, I was 15 or 16, and worked for Loew's Theater in Dallas. Mr. Hawkings was lecturing at El Centro College, about a 100 feet away. The Assistant manager there that night was John C. Cantu, still teaching Philosophy at SIU, Michigan or University of Michigan. He was 17 then. We carried Mr. Hawkings up a staircase to the projection booth. I really didn't realize how important of a chance meeting it was until much later in life. Imagine the questions I would love to ask now!
43 • Hot distro (by Ennio on 2024-01-17 22:09:08 GMT from The Netherlands)
Given that AV Linux has announced the passage to MX as its base, it really seems MX Linux is quite hot...
44 • Nothing to Hide (by Nathan on 2024-01-17 23:00:30 GMT from United States)
I don't want Google to know what board games I like, not because I'm ashamed of my taste in games, nor because it isn't anybody's business (I freely share my gaming interests with my friends so that they'll come over to play). It's because I don't want Google crafting a browsing experience for me centered around manipulating me into purchasing yet another Catan expansion.
We have different levels of privacy for different people. My wife and kids know my daily schedule, and that's great (even necessary). What business does Apple, Amazon, or Google have in knowing such a personal detail? Actually, quite a bit of business - there is a lot of money in giving me well-timed advertisements for places I'll be driving past within the hour. Maybe some people aren't bothered by their personal lives being mined for marketing purposes, but I'm not one of those people.
45 • Voice Assistant and privacy (by Gary on 2024-01-18 00:27:11 GMT from Canada)
After installing Mycroft on ancient PC running MX Linux 21, my experience was of limited success, mostly due to incompatible sound card and it's drivers! I've no reason to doubt it would have worked on their hardware, but oddly, owners of their hardware seemed to keep pretty quiet about it before this project faded and transitioned to another vendor with interest in developing that OS.
Glad to note some of that Mycroft talent actively working to improve 'Home Assistant' in their 2023 Year of the Voice. Kudos to Mycroft successors for continuing to develop their products. I get that many (most) of the readers here wouldn't trade their privacy for the supposed convenience of using your voice, however, when facing loss of vision coupled with the sky high cost of assistance tech (for example https://www.letsenvision.com/glasses) the possibility of affordable voice assistance puts privacy concerns on the back burner (unfortunately).
Several android apps are on the play store: Envision AI, Seeing Assistant, Seeing AI, Lookout, Lazarillo and I am grateful for all development to assist!
46 • Privacy & co. (by Sondar on 2024-01-18 15:50:57 GMT from United Kingdom)
So nice to be amongst so many knowledgeable folk who understand the importance of personal privacy, etc. More reasons that I never switch on (or even carry about!) any of the half-dozen mobile phones that kind and well-meaning friends and family have equipped me with! It's a dangerous enough world as well as greed for my dosh barely masquerading as giving information. Information gathering has become an obsession - just don't play!
47 • Offline text-to-speech (by Oskar on 2024-01-19 01:22:53 GMT from Japan)
@2 Since OpenAI made its Whisper TTS model free to use a number of open source projects have incorporated it, like self-hostable whisper-based TTS web portals. My favorite is SubtitleEdit (a windows GUI subtitle editor), which now can do local TTS and output as a subtitle file. I have recorded talking to myself during my commute, and then used SubtitleEdit to get my "stream of consciousness" transcribed completely off-grid, with good results. Offline TTS is a very useful tool to have.
48 • MX - Hot (by 142857 on 2024-01-19 02:34:10 GMT from Germany)
@43 Perhaps this may be an invitation to administer at lower levels - or to try to run the system as smoothly as possible, at the expense of resources? However, MX actually delivers a very good product, but this factor is actually a bit irritating. There are also many processes running in the background. When changing the kernel, however, I was able to determine differences in performance and the temperature also dropped about 5 ° C. However, the utilization of the RAM is limited - and the response time is also okay. Just keep an eye on the conky - and hope that the harmony will return. MX is able to built less hungry OS's, that's why I can not understand this behaviour.
49 • Do you use a virtual/voice assistant? (by James on 2024-01-19 12:16:28 GMT from United States)
No, unless I am listening to music my sound is turned off. Even back in my Windows days I shut down Clippy, what an annoying character.
50 • Clippy (by Friar Tux on 2024-01-19 13:11:24 GMT from Canada)
@49 James I loved Clippy. Actually, when I was on Windows 95, I had this program that had a butler walk into the screen from the lower right side to make announcements - "You have mail, sir." "There appears to be an error, sir." Unfortunately, that was too long ago for me to remember the name of the program. But it was fun.
51 • MS assistants (by Friar Tux on 2024-01-19 19:14:17 GMT from Canada)
@49 (James) Oh, and I also enjoyed Microsoft's Rover, Courtney, and Merlin (my favourite). I haven't really found anything comparable in Linux. Any suggestions, anyone?
Number of Comments: 51
Display mode: DWW Only • Comments Only • Both DWW and Comments
| | |
TUXEDO |
TUXEDO Computers - Linux Hardware in a tailor made suite Choose from a wide range of laptops and PCs in various sizes and shapes at TUXEDOComputers.com. Every machine comes pre-installed and ready-to-run with Linux. Full 24 months of warranty and lifetime support included!
Learn more about our full service package and all benefits from buying at TUXEDO.
|
Archives |
• Issue 1095 (2024-11-04): Fedora 41 Kinoite, transferring applications between computers, openSUSE Tumbleweed receives multiple upgrades, Ubuntu testing compiler optimizations, Mint partners with Framework |
• Issue 1094 (2024-10-28): DebLight OS 1, backing up crontab, AlmaLinux introduces Litten branch, openSUSE unveils refreshed look, Ubuntu turns 20 |
• Issue 1093 (2024-10-21): Kubuntu 24.10, atomic vs immutable distributions, Debian upgrading Perl packages, UBports adding VoLTE support, Android to gain native GNU/Linux application support |
• Issue 1092 (2024-10-14): FunOS 24.04.1, a home directory inside a file, work starts of openSUSE Leap 16.0, improvements in Haiku, KDE neon upgrades its base |
• Issue 1091 (2024-10-07): Redox OS 0.9.0, Unified package management vs universal package formats, Redox begins RISC-V port, Mint polishes interface, Qubes certifies new laptop |
• Issue 1090 (2024-09-30): Rhino Linux 2024.2, commercial distros with alternative desktops, Valve seeks to improve Wayland performance, HardenedBSD parterns with Protectli, Tails merges with Tor Project, Quantum Leap partners with the FreeBSD Foundation |
• Issue 1089 (2024-09-23): Expirion 6.0, openKylin 2.0, managing configuration files, the future of Linux development, fixing bugs in Haiku, Slackware packages dracut |
• Issue 1088 (2024-09-16): PorteuX 1.6, migrating from Windows 10 to which Linux distro, making NetBSD immutable, AlmaLinux offers hardware certification, Mint updates old APT tools |
• Issue 1087 (2024-09-09): COSMIC desktop, running cron jobs at variable times, UBports highlights new apps, HardenedBSD offers work around for FreeBSD change, Debian considers how to cull old packages, systemd ported to musl |
• Issue 1086 (2024-09-02): Vanilla OS 2, command line tips for simple tasks, FreeBSD receives investment from STF, openSUSE Tumbleweed update can break network connections, Debian refreshes media |
• Issue 1085 (2024-08-26): Nobara 40, OpenMandriva 24.07 "ROME", distros which include source code, FreeBSD publishes quarterly report, Microsoft updates breaks Linux in dual-boot environments |
• Issue 1084 (2024-08-19): Liya 2.0, dual boot with encryption, Haiku introduces performance improvements, Gentoo dropping IA-64, Redcore merges major upgrade |
• Issue 1083 (2024-08-12): TrueNAS 24.04.2 "SCALE", Linux distros for smartphones, Redox OS introduces web server, PipeWire exposes battery drain on Linux, Canonical updates kernel version policy |
• Issue 1082 (2024-08-05): Linux Mint 22, taking snapshots of UFS on FreeBSD, openSUSE updates Tumbleweed and Aeon, Debian creates Tiny QA Tasks, Manjaro testing immutable images |
• Issue 1081 (2024-07-29): SysLinuxOS 12.4, OpenBSD gain hardware acceleration, Slackware changes kernel naming, Mint publishes upgrade instructions |
• Issue 1080 (2024-07-22): Running GNU/Linux on Android with Andronix, protecting network services, Solus dropping AppArmor and Snap, openSUSE Aeon Desktop gaining full disk encryption, SUSE asks openSUSE to change its branding |
• Issue 1079 (2024-07-15): Ubuntu Core 24, hiding files on Linux, Fedora dropping X11 packages on Workstation, Red Hat phasing out GRUB, new OpenSSH vulnerability, FreeBSD speeds up release cycle, UBports testing new first-run wizard |
• Issue 1078 (2024-07-08): Changing init software, server machines running desktop environments, OpenSSH vulnerability patched, Peppermint launches new edition, HardenedBSD updates ports |
• Issue 1077 (2024-07-01): The Unity and Lomiri interfaces, different distros for different tasks, Ubuntu plans to run Wayland on NVIDIA cards, openSUSE updates Leap Micro, Debian releases refreshed media, UBports gaining contact synchronisation, FreeDOS celebrates its 30th anniversary |
• Issue 1076 (2024-06-24): openSUSE 15.6, what makes Linux unique, SUSE Liberty Linux to support CentOS Linux 7, SLE receives 19 years of support, openSUSE testing Leap Micro edition |
• Issue 1075 (2024-06-17): Redox OS, X11 and Wayland on the BSDs, AlmaLinux releases Pi build, Canonical announces RISC-V laptop with Ubuntu, key changes in systemd |
• Issue 1074 (2024-06-10): Endless OS 6.0.0, distros with init diversity, Mint to filter unverified Flatpaks, Debian adds systemd-boot options, Redox adopts COSMIC desktop, OpenSSH gains new security features |
• Issue 1073 (2024-06-03): LXQt 2.0.0, an overview of Linux desktop environments, Canonical partners with Milk-V, openSUSE introduces new features in Aeon Desktop, Fedora mirrors see rise in traffic, Wayland adds OpenBSD support |
• Issue 1072 (2024-05-27): Manjaro 24.0, comparing init software, OpenBSD ports Plasma 6, Arch community debates mirror requirements, ThinOS to upgrade its FreeBSD core |
• Issue 1071 (2024-05-20): Archcraft 2024.04.06, common command line mistakes, ReactOS imports WINE improvements, Haiku makes adjusting themes easier, NetBSD takes a stand against code generated by chatbots |
• Issue 1070 (2024-05-13): Damn Small Linux 2024, hiding kernel messages during boot, Red Hat offers AI edition, new web browser for UBports, Fedora Asahi Remix 40 released, Qubes extends support for version 4.1 |
• Issue 1069 (2024-05-06): Ubuntu 24.04, installing packages in alternative locations, systemd creates sudo alternative, Mint encourages XApps collaboration, FreeBSD publishes quarterly update |
• Issue 1068 (2024-04-29): Fedora 40, transforming one distro into another, Debian elects new Project Leader, Red Hat extends support cycle, Emmabuntus adds accessibility features, Canonical's new security features |
• Issue 1067 (2024-04-22): LocalSend for transferring files, detecting supported CPU architecure levels, new visual design for APT, Fedora and openSUSE working on reproducible builds, LXQt released, AlmaLinux re-adds hardware support |
• Issue 1066 (2024-04-15): Fun projects to do with the Raspberry Pi and PinePhone, installing new software on fixed-release distributions, improving GNOME Terminal performance, Mint testing new repository mirrors, Gentoo becomes a Software In the Public Interest project |
• Issue 1065 (2024-04-08): Dr.Parted Live 24.03, answering questions about the xz exploit, Linux Mint to ship HWE kernel, AlmaLinux patches flaw ahead of upstream Red Hat, Calculate changes release model |
• Issue 1064 (2024-04-01): NixOS 23.11, the status of Hurd, liblzma compromised upstream, FreeBSD Foundation focuses on improving wireless networking, Ubuntu Pro offers 12 years of support |
• Issue 1063 (2024-03-25): Redcore Linux 2401, how slowly can a rolling release update, Debian starts new Project Leader election, Red Hat creating new NVIDIA driver, Snap store hit with more malware |
• Issue 1062 (2024-03-18): KDE neon 20240304, changing file permissions, Canonical turns 20, Pop!_OS creates new software centre, openSUSE packages Plasma 6 |
• Issue 1061 (2024-03-11): Using a PinePhone as a workstation, restarting background services on a schedule, NixBSD ports Nix to FreeBSD, Fedora packaging COSMIC, postmarketOS to adopt systemd, Linux Mint replacing HexChat |
• Issue 1060 (2024-03-04): AV Linux MX-23.1, bootstrapping a network connection, key OpenBSD features, Qubes certifies new hardware, LXQt and Plasma migrate to Qt 6 |
• Issue 1059 (2024-02-26): Warp Terminal, navigating manual pages, malware found in the Snap store, Red Hat considering CPU requirement update, UBports organizes ongoing work |
• Issue 1058 (2024-02-19): Drauger OS 7.6, how much disk space to allocate, System76 prepares to launch COSMIC desktop, UBports changes its version scheme, TrueNAS to offer faster deduplication |
• Issue 1057 (2024-02-12): Adelie Linux 1.0 Beta, rolling release vs fixed for a smoother experience, Debian working on 2038 bug, elementary OS to split applications from base system updates, Fedora announces Atomic Desktops |
• Issue 1056 (2024-02-05): wattOS R13, the various write speeds of ISO writing tools, DSL returns, Mint faces Wayland challenges, HardenedBSD blocks foreign USB devices, Gentoo publishes new repository, Linux distros patch glibc flaw |
• Issue 1055 (2024-01-29): CNIX OS 231204, distributions patching packages the most, Gentoo team presents ongoing work, UBports introduces connectivity and battery improvements, interview with Haiku developer |
• Issue 1054 (2024-01-22): Solus 4.5, comparing dd and cp when writing ISO files, openSUSE plans new major Leap version, XeroLinux shutting down, HardenedBSD changes its build schedule |
• Issue 1053 (2024-01-15): Linux AI voice assistants, some distributions running hotter than others, UBports talks about coming changes, Qubes certifies StarBook laptops, Asahi Linux improves energy savings |
• Issue 1052 (2024-01-08): OpenMandriva Lx 5.0, keeping shell commands running when theterminal closes, Mint upgrades Edge kernel, Vanilla OS plans big changes, Canonical working to make Snap more cross-platform |
• Issue 1051 (2024-01-01): Favourite distros of 2023, reloading shell settings, Asahi Linux releases Fedora remix, Gentoo offers binary packages, openSUSE provides full disk encryption |
• Issue 1050 (2023-12-18): rlxos 2023.11, renaming files and opening terminal windows in specific directories, TrueNAS publishes ZFS fixes, Debian publishes delayed install media, Haiku polishes desktop experience |
• Issue 1049 (2023-12-11): Lernstick 12, alternatives to WINE, openSUSE updates its branding, Mint unveils new features, Lubuntu team plans for 24.04 |
• Issue 1048 (2023-12-04): openSUSE MicroOS, the transition from X11 to Wayland, Red Hat phasing out X11 packages, UBports making mobile development easier |
• Issue 1047 (2023-11-27): GhostBSD 23.10.1, Why Linux uses swap when memory is free, Ubuntu Budgie may benefit from Wayland work in Xfce, early issues with FreeBSD 14.0 |
• Issue 1046 (2023-11-20): Slackel 7.7 "Openbox", restricting CPU usage, Haiku improves font handling and software centre performance, Canonical launches MicroCloud |
• Issue 1045 (2023-11-13): Fedora 39, how to trust software packages, ReactOS booting with UEFI, elementary OS plans to default to Wayland, Mir gaining ability to split work across video cards |
• Issue 1044 (2023-11-06): Porteus 5.01, disabling IPv6, applications unique to a Linux distro, Linux merges bcachefs, OpenELA makes source packages available |
• Issue 1043 (2023-10-30): Murena Two with privacy switches, where old files go when packages are updated, UBports on Volla phones, Mint testing Cinnamon on Wayland, Peppermint releases ARM build |
• Issue 1042 (2023-10-23): Ubuntu Cinnamon compared with Linux Mint, extending battery life on Linux, Debian resumes /usr merge, Canonical publishes fixed install media |
• Issue 1041 (2023-10-16): FydeOS 17.0, Dr.Parted 23.09, changing UIDs, Fedora partners with Slimbook, GNOME phasing out X11 sessions, Ubuntu revokes 23.10 install media |
• Full list of all issues |
Star Labs |
Star Labs - Laptops built for Linux.
View our range including the highly anticipated StarFighter. Available with coreboot open-source firmware and a choice of Ubuntu, elementary, Manjaro and more. Visit Star Labs for information, to buy and get support.
|
Random Distribution |
linuX-gamers Live DVD
The linuX-gamers Live DVD was a bootable medium with a collection of popular 3D games. It contains both the NVIDIA and ATI proprietary Linux drivers and includes the following games: BzFLag, Enemy Territory, Glest, Nexuiz, Sauerbraten, Torcs, Tremulous, True Combat: Elite, Warsow and World of Padman. Although the live DVD was based on Arch Linux, it does not offer any desktop Linux functionality except gaming.
Status: Discontinued
|
TUXEDO |
TUXEDO Computers - Linux Hardware in a tailor made suite Choose from a wide range of laptops and PCs in various sizes and shapes at TUXEDOComputers.com. Every machine comes pre-installed and ready-to-run with Linux. Full 24 months of warranty and lifetime support included!
Learn more about our full service package and all benefits from buying at TUXEDO.
|
Star Labs |
Star Labs - Laptops built for Linux.
View our range including the highly anticipated StarFighter. Available with coreboot open-source firmware and a choice of Ubuntu, elementary, Manjaro and more. Visit Star Labs for information, to buy and get support.
|
|