Canada Post is a service, not a consumer product. The idea that it operates at a "loss" is stupid. It's like saying schools operate at a loss; hospitals operate at a loss; roads operate at a loss; sewers operates at a loss.

We pay for public services collectively, because they are simply needed for our society for function and for everyone to have (relatively) equal access to participate in culture/society/democracy.

Stop reducing necessary services to an income statement!
#cdnpoli #canadapost

Has anyone looked at whether using #AI for coding will hold back adopting of new (improved) language tools and standards?

For example, #Drupal has used annotations for plugins for years, while #PHP attributes are the new preferred syntax. AIs are trained off of code of the past, which for plugins was all annotations until attributes were introduced. Does that mean the stochastic coding parrot will continue to suggest annotations because of their stronger weight in the training set?

Do I have anyone in my network that works in IT for a school district? We want to start working on an Android replacement for EOL ChromeBooks and Windows devices, but we also want to make it so moving to the new OS is as lateral a transition as possible, so I am looking to work with individuals that can help enlighten me as to what features they use for device management and what features are required for all the classroom applications.

L’#écritureinclusive, une arme contre les #stéréotypes de #genre
theconversation.com/lecriture-…
L’écriture inclusive vise une égalité des représentations entre femmes et hommes. Mais face à la diversité des options proposées, quelle stratégie adopter ? Une étude récente montre que les formulations rendant visibles à la fois le masculin et le féminin – comme « étudiants et étudiantes » – sont les plus efficaces pour réduire les stéréotypes de genre.

A contrast in two announcements today. This is why OpenAI is a different kind of company than Meta and should not be distracted with an ad-based business.

ChatGPT Pulse from OpenAI. Fidji Simo:

"Pulse has already helped me discover new emerging treatments for my health condition, recommended new painting techniques for my art practice, surfaced great weekend events for my family, and more."


Vibes from Meta. Mark Zuckerberg:

"Introducing Vibes – a feed of expressive AI-generated videos from artists and creators in the Meta AI app."
This entry was edited (2 weeks ago)

My mom just locked herself out of her Mac after forgetting, incorrectly entering, mixing up the password. I hope it can get sorted out, but meanwhile... This is the fucking reason you use a password manager! Is it seriously, seriously that hard? I'm saying it for months now. But tbf, this is probably the way how I'd learn best as well.
in reply to André Polykanine

@menelion Well, yeah. I was rather trying to make the point that I find this pseudo security a bit over the top. In my school, people found it super funny to block people from using their iPads for a few minutes by entering the pin wrong. And well, if you do it long enough, your Apple device is basically bricked (unless you wanna lose the data). This can't happen on Windows (I mean you can always just connect the drive externally, or restore with Microsoft Account (very useful feature imo), and so on. But well, it is how it is. My point is it could've been avoided. And I've been mentioning the reason for months, if not a half year, with small to no interest. I'm just sorry for people apparently not getting how important this part is, and I knew that the day would come. And I'd say that is rather a mild case.

📣 The Jami team is proud to present Atlas: our biggest step toward reliable communication. 🌐⚡
Groundbreaking improvements to connectivity & delivery, even in complex networks.

🔗 Full article : jami.net/atlas-jamis-biggest-s…

#Jami #OpenSource #P2P #PrivacyMatters

This entry was edited (2 weeks ago)

ID cards are bad. Very bad.

1. Once they exist it will become necessary to always carry it. If you don't you must have something to hide, etc.
2. They can't stop people working in the 'black economy' because people currently employing those working for cash don't care.
3. It is another attack on trans and NB people (very unlikely someone would be allowed to have multiple IDs)
4. Massive data loss of personal information is highly likely.
5. Who pays? Why should they?
6. Police state becomes more likely with having your ID inspected whenever a copper wants to harass someone, especially POC.
7. OTHER OPTIONS ALREADY EXIST!

ID cards are a very bad, very dangerous, very expensive, and very risky idea.

ps I'm not a 'Brit'. I'm British, or more specifically English!
#IDcards #PoliceStateUK #BritCard

This entry was edited (2 weeks ago)
in reply to AlisonW ♿🏳️‍🌈♾️

I'm curious, as a Belgian with a mandatory ID card (and French, same situation), it's never felt like such a huge issue.
You can get your gender changed fairly easily and it is reflected on the ID card. and it makes identifying with online government services much easier. (think taxes, unemployment benefits, etc)
It has zero impact on taxes avoidance or anything like that but at least in Belgium it was never meant to so that's ok.
in reply to Edzilla

@edzilla I was part of the team that implemented the e-ID in Flanders.

The difference is that Schengen ID cards are just that. The provide an ID in the same way a passport does, and, because they are also a smartcard, they can be used not only for identification but also authentication.

Key is that it sits securely in your own pocket.

The proposed system is a central database with a viewer in the form of a mobile app. The opposite of "in your pocket", it's more like the eVISA system.

This entry was edited (2 weeks ago)

Vom 30. Oktober bis zum 02. November 2025 treffen sich die freien #Radios in Europas Kulturhauptstadt #Chemnitz für ein ganzes Wochenende. Das Kongressprogramm, von #Linux Audio über #inklusion in den Radios bis zur #medienpolitik, ist Online: community-media.net/programm-2…

Ein paar Beschreibungstexte fehlen noch, die kommen in den nächsten Tagen. Wir freuen uns über rege Teilnahme!

#zwcm2025

SAP software not suited for SAAQclic project, inquiry hears

ctvnews.ca/montreal/article/sa…

In 2000 I heard a consultant training on Sun sysadmin saying "You don't shut down a SAP service. You shut down the server".

It was already bloat.

in reply to Matt Campbell

nope. Nothing visual. The video is just a drummer playing at his kit.

It is annoying to jazz musicians because it is a completely inappropriate beat that you would never ever choose for jazz, or for that song.

It is humorous to music nerds with a rock background, in that we immediately recognize it as a simplified rock beat that we’ve heard countless times - a special note is the Bass drum - and also how much it mostly works despite how it really shouldn’t.

This entry was edited (2 weeks ago)

By teaching us to view poverty as a moral failure of the individual, society keeps us from realizing that poverty is actually an ethical failure of society itself. We can choose with #UBI to prevent everyone from ever experiencing the poverty that 60% of us will experience in life, and we just don't.

youtube.com/shorts/7AgzQuon63Y…

#ubi

Just one month until I launch my newest app on iPhone, iPad, and Mac. It's called Hacktivate, and it teaches real-world computer science skills using cybersecurity challenges. You can pre-order it here: apps.apple.com/gb/app/hacktiva…

Because it's 2025... shouldn't this been the case in 2005?!? montreal.citynews.ca/2025/09/2… #cdnpoli #polcan
in reply to I am Water

@SlicerDicer even though it doesn't have API spark ignition ratings on the bottle, it at least meets SJ as that's a prerequisite for the JASO MA/MA2 ratings that it has on the bottle 🤷‍♂️

looks like the anti-wear additive package is on the ragged edge of working with newer gasoline cars' catalytic converters though, so maybe don't actually run it in something very modern

that said I only actually use it in one of my engines, which is spark ignition, but has no catalytic converter and calls for JASO MA (it also doesn't actually have a wet clutch, so it really doesn't need that spec)

in reply to bhtooefr

@SlicerDicer basically the only actual spec I’m violating by running T6 5W-40 in a 2007 Honda Helix is, ironically, the oil weight spec

(the engine came out in 1984 and called for 10W-40, then in 2006 Honda retroactively changed all the oil specs to 10W-30. I’ll follow that if the engine was designed for it, but this one wasn’t. so, 5W to get faster cold start lubrication, and 40 to go for the original specs for high temp lubrication.)

meanwhile my other shit actually gets what the manufacturer calls for, which is not a diesel oil. my Prius gets the 0W-20 that it calls for because the cold performance is critical in a full hybrid, my Honda NC700X gets a 10W-30 motorcycle oil (and Honda’s 10W-30 is almost a 10W-20) partially because it has a DCT with hydraulic clutch actuation using the engine oil. (I wish T6 Multi-Vehicle 5W-30 were JASO-rated though, going down to a 5W might actually help with some clutch grabbiness issues I have when it’s cold, and I think the JDM oil Honda sells for these bikes is a 5W-30 now, too.)

Ok. We're going to attempt this Linux Gnome3 upgrade again this afternoon. This time, I'm getting AT-SPI 2.6.2. What did change in GNOME 40+ (and carried forward into GNOME 42, 43, … up to Trixie’s GNOME 45) is how tightly Orca and the Shell depend on the D-Bus accessibility bus (at-spi2-core), and the versions in play. Debian Trixie ships at-spi2-core 2.56.x (part of GNOME 45 stack). Upstream today is around 2.62.x (tagged for GNOME 47). Orca 49 is written to work with both, but in practice some focus events and menu roles are only exposed properly with the newer stack.
So if I had Orca 49 + GNOME Shell but only the 2.56 accessibility stack, that mismatch could explain why menus and panels were "dead air" - the Shell wasn’t emitting the AT-SPI events Orca was waiting for. MATE, on the other hand, doesn’t rely on the GNOME Shell introspection path, so Orca 49 + AT-SPI 2.56 works just fine there.
python3-pyatspi is much newer in Forky (2.57.0-1 vs 2.46.x in Trixie).
at-spi2-core/libatspi2.0-0/at-spi2-common: Forky currently tracks 2.56.2-1, which matches or slightly bumps it, but I'm afraid, it won't be enough for Gnome latest. OK, Trixie-backports has 2.5.8 in it, much better.
Boy am I learning a lot about Linux today! :D
This entry was edited (2 weeks ago)
in reply to feld

example:

# smartctl -a /dev/nvme0
smartctl 7.5 2025-04-30 r5714 [FreeBSD 14.3-RELEASE-p2 amd64] (local build)
Copyright (C) 2002-25, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Number: Samsung SSD 980 PRO 2TB
Serial Number: S6B0NG0R604351N
Firmware Version: 5B2QGXA7
PCI Vendor/Subsystem ID: 0x144d
IEEE OUI Identifier: 0x002538
Total NVM Capacity: 2,000,398,934,016 [2.00 TB]
Unallocated NVM Capacity: 0
Controller ID: 6
NVMe Version: 1.3
Number of Namespaces: 1
Namespace 1 Size/Capacity: 2,000,398,934,016 [2.00 TB]
Namespace 1 Utilization: 1,068,887,175,168 [1.06 TB]
Namespace 1 Formatted LBA Size: 512
Namespace 1 IEEE EUI-64: 002538 b61150404c
Local Time is: Thu Sep 25 11:15:57 2025 PDT
Firmware Updates (0x16): 3 Slots, no Reset required
Optional Admin Commands (0x0017): Security Format Frmw_DL Self_Test
Optional NVM Commands (0x0057): Comp Wr_Unc DS_Mngmt Sav/Sel_Feat Timestmp
Log Page Attributes (0x0f): S/H_per_NS Cmd_Eff_Lg Ext_Get_Lg Telmtry_Lg
Maximum Data Transfer Size: 128 Pages
Warning Comp. Temp. Threshold: 82 Celsius
Critical Comp. Temp. Threshold: 85 Celsius

Supported Power States
St Op Max Active Idle RL RT WL WT Ent_Lat Ex_Lat
0 + 8.49W - - 0 0 0 0 0 0
1 + 4.48W - - 1 1 1 1 0 200
2 + 3.18W - - 2 2 2 2 0 1000
3 - 0.0400W - - 3 3 3 3 2000 1200
4 - 0.0050W - - 4 4 4 4 500 9500

Supported LBA Sizes (NSID 0x1)
Id Fmt Data Metadt Rel_Perf
0 + 512 0 0

=== START OF SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

SMART/Health Information (NVMe Log 0x02, NSID 0xffffffff)
Critical Warning: 0x00
Temperature: 30 Celsius
Available Spare: 96%
Available Spare Threshold: 10%
Percentage Used: 91%
Data Units Read: 3,206,095,844 [1.64 PB]
Data Units Written: 2,044,433,376 [1.04 PB]
Host Read Commands: 62,454,506,739
Host Write Commands: 18,036,238,728
Controller Busy Time: 140,409
Power Cycles: 83
Power On Hours: 36,312
Unsafe Shutdowns: 48
Media and Data Integrity Errors: 569
Error Information Log Entries: 569
Warning Comp. Temperature Time: 0
Critical Comp. Temperature Time: 0
Temperature Sensor 1: 30 Celsius
Temperature Sensor 2: 37 Celsius

Error Information (NVMe Log 0x01, 16 of 64 entries)
No Errors Logged

Self-test Log (NVMe Log 0x06, NSID 0xffffffff)
Self-test status: No self-test in progress
No Self-tests Logged

in reply to André Polykanine

I just changed it to a Python text to speech library instead to make it less Linux focused. I think the library simply calls Linux's Espeak or Window's Speak.
The commadline text part should have worked.
It does take an LLM backend though, I use llama.cpp. I haven't tested it with ollama yet.
If you wanted chatGPT to do it we could probably add a spot for an API key. I'm normally more focused on using local models.

Also, no way to use NVDA Remote client servers on Linux either. I'm not even wanting to host one, just connect to an NVDA client and control it from within Linux. I'm afraid this is not yet a thing, which is a bit of a shame considering iOS, Android all have NVDA Remote protocol apps. I need this for my work machine, I often use a local remote server over internal ethernet to wire up my work and leasure machines to control them from within one keyboard and headset. Linux won't really be a viable OS for me until I can do this again.
(update): Thanks to @menelion for mentioning this project - this is promising, reach out to the author for questions. github.com/gozaltech/nvdaremot…
This entry was edited (2 weeks ago)
in reply to Matt Campbell

@matt I don't know of a Mac OS one, but do know that there was an iOS one which seems to have been abandoned. I'm not sure what's going on with it, or why the code was never released, but it may be worth asking the author and seeing if there's any answer github.com/EscolarProgramming/… For windows, there is, of course, the one which was mentioned earlier here github.com/gozaltech/nvdaremot… There's also the android client, the author says he got sick of Google being Google and decided not to bother any longer nvda-addons.groups.io/g/nvda-a… it may be worth seeing if the code is available.
in reply to Tamas G

There's always www.virtualhere.com and their modular KVM thing. It does work as an input source, and obviously works with any keyboard which supports multiple dongles (Logitech or handheldsci.com or whatever). The modular KVM thing works, too, it gives you a Window to the other machine which works well modularkvmip.com but I've only used it on Windows. I've never used it on Linux. I have used the virtualhere client itself on a RPI, but obviously that's just a keyboard/keyboard channel for the Windows machine. All this has nothing to do with sound, the only thing I can think of is Sonobus for very low latency and the program @menelion for NVDARemote speech. As for the mobile clients, are they really usable? They're great to have, but only for emergencies, in my experience.