The West Bend QuikServe Toaster browns bread in less than two minutes, one of the fastest toasters on the market. Items are held to toast, then released to bottom serving tray. Toast, bagel, and gluten free settings.
The rise and fall of Stack Overflow is a case in point of the parasitic nature of LLMs. LLMs feed their models on places like Stack Overflow to be useful to users, so users flock to them to avoid the eternal snarky comments and just get an answer to their problem right away. But this is a dead end. No new answers will be generated if no one uses Stack Overflow or similar places.
What goes for Stack Overflow goes essentially for the whole internet. Like a mold growing on food, consuming it, and dying once the food is gone - LLMs will kill large parts of the 'old' internet before long.
Men. Unless you're in the army, there's no reason to wear camouflage. None whatsoever. You don't look cool, you don't look tough, and most of the time you don't even look like a tree.
@matt Not exactly, but it seems that we need to be able to have it built for multiple Python flavors on openSUSE/SLE, and the examples that I'm finding for doing that use wheels, and, in theory, it shouldn't take too long, and I need to make a new release anyhow.
Well over lunch, I discovered something funny. ChatGPT has access to Espeak inside its own sandbox environment, so it can run Espeak and generate IPA forms for what Espeak would output. Huge help. So, we fixed US English, next up is patching other languages. Both UK and US English now sound like they should in NV Speech player. Add-on link: eurpod.com/nvSpeechPlayer-2025… (or original link works too, always latest there if you have a nice Browser like Firefox that doesn't cache downloads.) Oh, and it also adds a dummy normalizer for empty IPA phonemes (that have no mapping right now.) We send a dummy 10-frames of audio in that case so speech shouldn't stall during say-alls.
ahaha, no shit Sherlock. "I found that the espeak command tries to open audio and prints ALSA warnings. To suppress these, I could try --stdout or -q for quiet mode, but it still may attempt audio. To only get phoneme output, I can use 2>/dev/null to ignore warnings and suppress audio. Let's test this approach."
@maysoncabo I wonder if we can get the intonation format from the old commit, ee80f4d, working in this one? In my opinion that was when SpeechPlayer sounded the best, it had more of an Eloquence intonation sound whereas this one sounds more like eSpeak.
sort of. give eurpod.com/nvSpeechPlayer-2025… a try - It attempts to bring back some of the smoothness that was around from that commit, but we can't quite bring back the same exact way of frame generation as a lot changed + a pitch setting added to frame generation that fundementally makes it incompatible to revert with the newer driver.
@BTyson @maysoncabo nah, but I think introducing US English letters changed some of the British ones, so the IPA tokenizer also needs to know the nuance between letter types per language and how ESpeak represents that data. Bleh. Sadly that DLL change wasn't much, it did make S's a bit softer which is probably enoough of a downside to not keep it for now anyway. But yeah, really don't know what beyond the ends-of-pitches clipping they did different in their inflection logic, but the entire player itself changed and that may have been enough to offset inflection differences too.
@maysoncabo There was one commit where Mick said he totally redid the intonation system to be more like eSpeak's and from then on I wasn't as much a fan of SpeechPlayer for some reason.
> Nach § 124 OWiG ist das „unbefugte“ Verwenden dieser Symbole eine Ordnungswidrigkeit. Diese wird jedoch, insbesondere im Umfeld von Sportveranstaltungen, meist nicht geahndet.
i think actually displaying this anywhere in public would indeed be an Ordnungswidrigkeit? If anyone would ever do something about it is a different question. (And well, Ordnungswidrigkeit might also mean, it costs 5€ 🤷♀️).
Can someone with an HN account get some eyes on this project: r.lily-is.land/infra/lilybuild ? It might actually get an influx of support if people know it exists.
You can run Gitlab CI jobs from Gitea/Forgejo using Podman instead of Docker.
So you don't have to suffer with the Github Actions junk...
Really digging this fastSM client. The only thing I wish for BlueSky is for the ability to view feeds, but perhaps that's hiding somewhere and I've not found it. That and, this thing is new. I don't expect Egypt to be built overnight.
Catima is a simple Android app to store event tickets/plane tickets/store discount codes and more in.
Version 2.41.5 is the last iteration of the barcode encoding feature. It removes automatic encoding in favour of deciding encoding once on adding a card (extracted from pkpass files, otherwise ISO-8859-1) and then sticking with that. It also fixes some other bugs.
Fix list widget sometimes opening wrong card
Fix several bugs with shortcut handling
Fix About activity not using pure black title bar in OLED mode
Fully remove automatic barcode encoding detection...
@hipgnose Can't wait to see the lockscreen widget experience myself :) Maybe you have something useful to add to github.com/CatimaLoyalty/Andro… btw? Idk if I am missing something stupid or Google just didn't expose this feature in emulators again (they've done that before) :)
Hi! I was beginning to set up my lockscreen widgets (recent addition to android), and it seems Catima has one to provide although it's currently quite limited since currently: it only provides a co...
THERE IS A WAY TO NULLIFY AND STOP IT. Please read.
No promotions, no affiliation, not sponsored. I just use this service myself - they have no idea who I am. Please, I'm saying this to protect people I'm not benefiting from this post.
ICE with Webloc is able to scan entire neighborhoods and obtain mobile phone service data. On of the most critical pieces of information they log is your IMSI. Its like your car's VIN number, it identifies that device uniquely, and is the identifier the cell tower sees when your phone connects to it.
Now your IMSI and its location is logged, and cataloged to track your location over time; where you work, hangout, live, etc.
YOU NEED TO ROTATE YOUR IMSI TO MAKE THE DATA USELESS, and rotate it often.
My cell service provider does this automatically every 24 hrs. I only know of 1 cell provider that does it = Cape.co. There are other privacy focused providers, but they focus on minimal data collection (Cape does that too) and other security practices first.
Cape:
Rotates your IMSI every 24 hours (or you can refresh it manually)
Prevents man-in-the-middle attacks and Stringrays with Enhanced Signal detection. If the cell tower your phone attempts to connect to does not match to a known CAPE middle core, it will not connect. It also alerts in the app when suspicious connections are attempted and automatically blocks them.
Collects 0 PII - not even your billing info. They known the phone number they are servicing and that's it - not your name, address, DOB, nothing. If they receive legal demand to hand over your data they cannot, as they don't know who you are and never logged it in the first place.
Normally I'd include a coupon here, but to show you I'm extremely serious, I'm not going to include it. If you really want $10 off for life you can find a coupon somewhere, and they have a promotion rn, but I will not directly benefit from this post.
ok so this is built on "US Cellular's IoT-M network" (whatever the heck that means), and USCellular is now T-Mobile... so unless this requires some super special towers you should have great coverage anyway
if you choose to use this service you should probably get a completely new number and not port your number or they could probably still identify you with existing data they've collected associated with that number after they subpoena Cape to know which number is associated with the IMSI at a point in time
Der Umzug ist erfolgt. 💪Noch kommen Benachrichtigungen bei Misskey, doch reagieren und Antworten geht nicht mehr. Also nicht wundern. 😉#Sharkey#MisskeyDe#Umzug
Minnesota is really putting the facts of the 'opposition' party on show.
Walz says "I'm gonna deploy the National Guard," but actually deployed the state troopers to protect the Gestapo.
Now Frey's up on stage saying that MPD is just too busy protecting the economy to deal with crimes against humanity, so all protesters should go home. Because MPD won't protect them.
This is not opposition; it's collaboration and gaslighting.
he was never going to deploy the National Guard to help the protesters anyway; it would have been to protect building/property from a riot. He can't do anything with the National Guard anyway because Trump can instantly Federalize them.
The IMAP response parser would incorrectly detect {size} literals
inside quoted strings, causing protocol desynchronization.
For example, a server response like:
* 1 FETCH (BODY[HEADER] "Subje...
It seems that py/cryptography's thoughts about OpenSSL (cryptography.io/en/latest/stat…) are doing the rounds at the moment.
I've not touched OpenSSL directly in a long time. In fact, it appears that the 10-year anniversary of that (imperialviolet.org/2015/10/17/…) passed by a few months ago!
So I've no direct comments on the piece but, a long time ago, I was in the position where I was landing changes in both OpenSSL and NSS (Mozilla's TLS library). OpenSSL was somewhat famous for having bad code. And, indeed, if you looked at it back then the functions were full of single-letter variable names with pointer arithmetic everywhere and context-free, somewhat scary comments. It wasn't outside the norm for 1990s C code, but I understand why people recoiled.
In contrast, if you looked at NSS code, it looked good! Consistent formatting (before clang-format), good naming, good comments.
But NSS had a PKCS#11 abstraction layer and, even after years, I never could understand how the control flow worked there. I would have to single-step in gdb every time to figure out where an operation grounded out into actual code. I was reminded of that when reading py/cryptography's descriptions of OpenSSL 3.0.
I had a pet theory at the time that, because OpenSSL was repulsive on the surface, it inhibited people enough that they couldn't add much deeper complexity. But NSS, with its invitingly clean-looking code, was understandable and then people had enough capacity left over to add deeper complexity.
There might be something to it, although you shouldn't discount the fact that entities who are willing to fund cryptography libraries often have demands that are contrary to clean code. Things like FIPS compliance and compatibility with a zoo of different accelerators and bespoke needs.
So rather it might have been that old OpenSSL was old OpenSSL because it was mostly unfunded. That meant that it looked pretty ragged, but also there weren't so many demands in tension with good design.
NSS was funded by interests that really cared about PKCS#11 compatibility so that you could use a super-expensive, certified-everything HSM with it. When OpenSSL got shocked into switching to a higher-funding model, that brought lots of those same sorts of competing interests, and then the incentives pointed towards adding slow, impenetrable layers of abstraction all over.
So @lwn is currently under the heaviest scraper attack seen yet. It is a DDOS attack involving tens of thousands of addresses, and that is affecting the responsiveness of the site, unfortunately.
There are many things I would like to do with my time. Defending LWN from AI shitheads is rather far from the top of that list. I *really* don't want to put obstacles between LWN and its readers, but it may come to that.
Attached: 1 image
Kudos to Robert Nagy (robert@). Without much fuss, he committed OpenWV and enabled Widevine support in Chromium. Now we can all enjoy Netflix, Disney+, and other DRM content on #OpenBSD.
https://marc.
@Bri Oh I'm sure you will. Some of us have been asking for quoting on this platform for a long time, so it's nice to see someone attempting to allow us to actually do it.
I fired up Windows long enough to try out FastSM. seems cool so far. I'll have to test Bluesky functionality since I'm hardly over there and don't have very many followers. I'm austinnix123@bsky.social.
I use this all the time on the RNIB Talking Book site. It makes browsing tables of search results much easier and it's a service we use quite a lot. Quoting Freedom Scientific / Vispero: JAWS Feature Spotlight: Smart Navigation
Want a faster, more intuitive way to move through web pages, PDFs, and HTML content? Smart Navigation helps you browse more efficiently by letting you navigate by controls, tables, or both, so you can focus on the information that matters.
In this archived 20 Minute Tech Tips episode, you’ll learn what Smart Navigation is, why it matters, how to enable it, and how to switch modes on the fly.
In this episode, Freedom Scientific trainer Ron Miller dives deep into JAWS Smart Navigation, a powerful feature designed to make navigating web pages, PDFs, and HTML documents more intuitive.
@Bri Actually I manipulated that manually, so it wasn't an official quote. It was a copy and paste job from the original post. If I were to officially quote it using FastSM, I suspect it would not read out.
@Bri Oh I'm sure you will. Some of us have been asking for quoting on this platform for a long time, so it's nice to see someone attempting to allow us to actually do it.
So now Techno DJs start remixing AI songs, as there was no other music? Ok tbf probably nothing new, but one of my favourite did, and it pisses me off because heck the quallity is bad. Like at least put effort into it rather than just taking the 1st result, slapping some base on it and calling it a day. I'm coming more and more to the opinion that AI music is just trash no matter what. I mean it's fascinating, like, the fact that's even possible. But I dislike it.
Many times during my career, I have heard that #Linux has no chance on the #desktop unless some corporation picks it up and prepares it for the masses. #ChromeOS was given as an example.
I think it is safe to say today that all those small Linux vendors and communities have beaten #Google with its Chrome OS.
@pemensik That's why I said "small Linux vendors and communities". Red Hat or Canonical are small vendors compared to Google. Red Hat's annual revenue is what Google makes in one week.
There also seems to be a large tick in Unknown, up to 16% from 11.2% in October (when Windows 10 went end of life.) I bet a lot of those are new Linux users too. One can hope, at least. gs.statcounter.com/os-market-s…
I have been selling computers at retail stores for the better part of 25 years. In the last 12 months is the first time I have had people come in to the store and ask if "Linux" were on any of my computers.
it's not necessarily a general purpose OS, but I suppose Valve works on SteamOS, might also be given as an example of a corporation preparing the Linux ecosystem for the masses.
Hey! Guess what peoples? I almost have 12000 posts. That's a lot of posts! I've accumulated all of this in a little over 3 years. This means that, on average, I post a bit less than 4000 posts a year, provided I did the math right, which I probably did not. If someone wants to do more accurate math, I believe I joined this instance on October 30, 2022. Or something like that. So have fun with that if you like. Lol
It's really funny to me how the amount of autism and dry humor in our family means that we will get into a big group text conversation and someone says something and then one of us is like, "Wait, is he kidding?" And then the other autistic person is like, "I think so?" And the first autistic person: "Oh, I knew that?"
And I literally can't tell who is being sarcastic or what
(but it's okay because everyone is nice - unlike growing up, when missing sarcasm meant DOOOOM, heh)
Since 2013, the IRS has released data culled from over 1.8 million nonprofit tax filings. Use this database to find organizations and see details like their executive compensation, revenue and expenses, as well as download tax filings going back as f…
This is the second layoff at Mozilla this year, the first affecting dozens of staff on the side of the organization that builds the popular Firefox browser.
Licensing revenue (what is paid by MoCorp and MZLA for using the Firefox and Thunderbird brands) grew by $3 million, so it's quite likely that MoCorp result was much better. But we'll never know the details if Mozilla doesn't release the financial report.
As for executive compensation, Mark Surman (president) got a small cut, while Mitchell Baker, MoCorp CEO had a 6.2$ million compensation.
This is either an error or a very nice salary, as she "left" the CEO role in February 2024.
Note that the compensation of the MoCorp CEO doesn't come from the MoFo, it's just mentioned in their report.
Also the foundation laid off 30% of its employees in November 2024 (advocacy division). This would remove around $5 million from the MoFo expenses.
Even though the MoFo has lost $4 million in 2024, they still had $117 million in assets at the end of the year. So they could have paid those laid-off workers for 23 years with those assets… Again, making the workers pay for execs errors.
feld
in reply to feld • • •there is no such thing as a slice of bread too wide to fit this toaster
you want to toast burger buns? hotdog buns? done. And they don't get stuck inside either.
≠ HER PENIS ≠
in reply to feld • • •westbend.com/collections/toast…
Looks promising
West Bend QuikServe Compact 2-Slice Toaster with Extra Wide Slots and Cool Touch Exterior, in Black (77224)
West Bendfeld
in reply to ≠ HER PENIS ≠ • • •