You're supposed to be able to do that, but even when I use a proper USB-C data cable with fast transfer speeds and look at the hardware tab, it says connected at 480 Mbps, instead of 5 or 10 Gbps.
Mastodon, conéctame con personas que les guste: - Ciencia de datos - Optimización Matemática - Investigación de Operaciones - Desarrollo de Software - IA/ML
when I agree to some (kinky) playtime with someone, I consent to playing/hurting them/tying them/whatever as long as it's ultimately for their enjoyment or fulfillment
not enjoying some act or dynamic and still not communicating it to your play companion, out of fear of disappointing them or not meeting their expectations, is a big betrayal of trust
An email states: "Rethink your customer journeys with us."
What I think that means is include us in your plans, but what it actually says to me is, ya know, you should really rethink using us in the first place.
Ukraine has a new strategy that takes the war to Russia. By striking Russian factories with long range weapons, Ukraine is systematically dismantling the Rus...
You know how billionaires end up with severe cognitive deficits as a result of becoming surrounded by yes men who constantly tell them their every idea is genius? What if we made a bot that just does that to everyone. I think that would be great.
Just having more code is like making a poem longer. It doesn’t necessarily make it better. It doesn’t gain more emotional impact to take a good poem and add to it.
Does Windows work 30% better or faster? No, of course not. I would be impressed if they used AI to make the system 30% /smaller/ or otherwise more efficient, but these folks don’t understand systems engineering, just grifting consumers with vendor lock-in.
With @icing's latest perf tweaks for HTTP/3 in curl, my theoretical max speed is now 1550 MB/s on my machine (at 100% CPU) with h2 doing 2464MB/s and h1 at 3303 MB/s
When downloading 100MB chunks 50 times in 50 parallel downloads.
(I say theoretical because the servers can't quite keep up with the client on localhost.)
To clarify: the nghttpx gets the data from an apache backend. So part of the 100% goes into that, meaning QUIC cpu use of nghttpx and curl is probably on par.
Whoever at MS decided that encrypting drives behind Windows 11 Home users' backs and not disclosing that, and also whoever at MS decided that confusing OneDrive sync with anything approaching a valid backup should lose their data and then step on a well-placed LEGO in the middle of the night
Highly irresponsible and their confusion-enabling has caused permanent data loss for people who didn't know they had a key they needed BEFORE losing access to their MS account
I was in this situation 4 months ago. After windows update and restart I was no longer able to boot and I was asked for that key during recovery. Fortunatelly I had access to the account on another device but it was frustrating.
> “We think we’re on the cusp of the next evolution, where AI happens not just in that chatbot and gets naturally integrated into the hundreds of millions of experiences that people use every day,” says Yusuf Mehdi, executive vice president and consumer chief marketing officer at Microsoft, in a briefing with The Verge. “The vision that we have is: let’s rewrite the entire operating system around AI, and build essentially what becomes truly the AI PC.”
So: What would it take to start the Linux accessibility revolution? Or maybe turn it from a consistent chatter among a medium-volume minority into a matter of urgency?
I know, realistically, that stripping all the garbage off our Windows PC will be easier than trying to make Linux work.
> “All the data that we see is when people use voice, they love it,” says Mehdi. Some of that data is the billions of minutes that people spend talking in Microsoft Teams meetings. “They’re talking through their computers today, and I think this change to ‘talk with and talk to’ will come to reality and we’ll see this thing really take off,” says Mehdi.
Oh, yes. Everyone knows that talking to an actual person is exactly the same as talking to a suphisticated nonsense generator that is baked into Windows itself.
I know that, just as ChatGPT can be useful sometimes, a voice-controled computer agent *could* help some people. That is, it could help people if it works. Remember that thing I posted like an hour ago? Imagine what could happen if your voice agent hallucinated your entire OS. At least when I was talking to GPT, I was the one controlling the computer.
Accessibility considerations aside, language like "Rewrite the entire operating system around AI" is pretty fucking unambiguous, and I'd love to hear from even one person who thinks this sounds like a good idea.
I'll wait.
> “We want every person making the move to experience what it means to have a PC that’s not just a tool, but a true partner,” says Mehdi.
Absolutely big shameless plug, but hey, someone's gotta be shameless in life. Today's podcast episode was talking about AI and the influence it's having on everyday well, everything. From media to interactions, magazines to videos etc, and of course music.
@FreakyFwoof I need to stick notifications on the podcast account. Mastodon notifications don't show reposts, so I'm missing them now. Definitely gonna go listen to that. Plug much appreciated.
holy crap this entire article is dillusional. Have you actually tried to use Copilot vision? For anything? It is bad. It does not work. In every instance I tried to get it to help me with anything at all, it has not worked. This might sound like hyperboly but I'm not kidding. Describing screenshots is about a million times more reliable than any of the live crap. And they know it! They know it's bad! They literally spell it out in the article! For fucks sake.
@talon Did you read the post I made this morning? I played around with it for hours ¥yesterday, and even the things it did well weren't enough to make up for the serious problems I encountered.
Hacker gets annoyed at Amazon’s Kindle apps, reverse-engineers the Kindle web reader’s protocol (which basically sends each page as a set of glyphs in a deliberately broken variant of SVG). Such obscurity, much security.
When not using Google Play services (e.g. #GrapheneOS, #LineageOS users), #Signal can be a real battery drain. @mollyim with @unifiedpush on the other hand is extremely battery efficient.
I finally started using Molly to send and receive messages via Signal. Molly is a hardened fork of Signal for Android, offering features such as an encrypted message database, automatic locking, shredding no longer needed secrets from RAM, notificati…
@ratcatcher What do you mean by 'directly'? Without MollySocket? Delivering push messages directly to WebPush endpoints is something the @signalapp server's would have to support. And Signal famously hates infrastructure that's not run by tech oligarchs.
@ratcatcher @signalapp It’s a bit complicated but broadly speaking both the Mastodon server and the Matrix server have support for an internet standard called 'WebPush'. That’s where the server HTTP POSTs the push notification to, and that’s how the notification chain gets started that eventually ends up on your phone. But the Signal server doesn’t support that standard.
@ratcatcher For the client that is true. (And Molly does for example talk directly to #Conversations_im). However the problem arises when the server sees a new message it needs to do 'something'. Like it needs to tell 'someone'. And that something to someone needs to be standardized.
In the animation on unifiedpush.org/developers/int… the top, purple arrow. That’s WebPush. And Signal doesn’t do that. And then MollySocket is a translation proxy for that.
Oh that’s the part where it’s probably better for @mollyim to take over this conversation. But yes, it needs to run on a server somewhere. Running it locally would probably be fairly resource intensive. The goal of MollySocket is more or less to shift that resource consumption to a server. If you were running it locally you might as well not use UP at all.
Traditionally, there was the assumption in #Christianity that people who did not get baptized would not go into heaven.
And based on my readings of Austrian #folklore , the notion that newborn children might die without even receiving an "emergency baptism" by a midwife was a major source of trauma to parents, since even those innocent souls would technically not be able to get into Heaven. And folklore would fill in the gaps left by Christian theology.
For instance, some tales claimed that the souls of such children would be led into a beautiful garden by Perchta (one of the incarnations of the female spirit/goddess Hulda/Holda/Holle). There, they would play for all eternity, and even after Earth was destroyed on Judgment Day, this garden would remain.
There is even one tale where a "miracle doctor" promised to use his quackery to reanimate the dead infant for just long enough that it gets baptized - but instead, Perchta shows up and made clear that she has already claimed the child's soul.
In other, darker tales, these souls would transform into black night birds and accompany the Wild Hunt on their travels.
It's a fascinating subject, and shows how folklore spins its own narratives that go beyond official doctrines.
Detective Petrov's eyes swept the cluttered room. "So," he said, "Leonid Chekhov, age 56, an automobile mechanic - found strangled in his study. The question is, who would want him dead?"
"Sir," said officer Gurin, "look - there's a pistol on the desk."
Detective Petrov stared at the weapon, frowning. "Do you realize what this means?" he asked. "Before we're don here, I have to fire this gun...."
Could someone who uses a text reader tell me if when I post a link and it throws up an image do you see a blank image or does it not register the link image at all. When I post a link and it throws up an image I try to give a link image description in my post. It’s usually the header image of the linked article & I check to see if the article gave alt text for the image. If they did, I just note that. Thanks #AltText #Alt4You #TextReader .
@FreakyFwoof Does it make sense if at the bottom of my post I just write “link image description” and then give the description? When I boost a post with a link image I do #Alt4you and try to add that it’s a link image I’m describing.
@FreakyFwoof The link previews don’t have an alt-text function that I can find so I just try to make a note that what I’m describing is a link preview image. I guess my question is does a text reader say that there is a link preview image but no description. Or does it not even tell you that there’s a link preview. I think I’ll keep doing it cause if it was me I’d want to now what image the link was throwing up.
Aaah I see now. Yeah, with a link it could go anywhere, so it's always nice when someone gives a bit of an overview of where it's going, I do the same. I see a lot of just raw links posted and I won't click it, even youtube videos. If I don't know what it's about, why am I going there in the first place?
Máte doma v šuplíku telefon, který už nepotřebujete? Doneste ho na #OpenAlt stánek mobilního linuxu a pomůžete jeho vývoji! Pomůžou nám následující, třeba i poškozené (kromě zákl. desky), modely:
Pixel 3 (i XL), Pixel 3a (i XL), Pixel 4a, Pixel 6, OnePlus 6, 6T, Sony Xperia XZ2, XZ3, Xiaomi Mi Mix 2S, 3, Pocophone F1, Mi 8 (i Pro, Explorer), LG G7 ThinQ, V35 ThinQ, Fairphone (jakýkoliv), SHIFT (jakýkoliv), Chromebooky (různé) Motorola Moto (různé) Samsung Galaxy (různé) Xiaomi (různé)
As we at Cloudron rely on both @kde and @gnome everyday, we took the opportunity now with KDE's 29th anniversary fundraiser to start donating there also. For gnome we already did.
Such fantastic projects, hard to imagine a life on Linux without them. Even harder to decide which one to use 😃
This week is KDE’s 29th anniversary. It may not be a nice round number like 25 or 30, but whenever another birthday rolls around for an independent project the size and scope of KDE — powered by the goodwill of its contributors and users — that’s rea…
Pátek, začíná mi pracovní “týden”. A zrovna dnes mě skolil nějaký moribundus, tak se láduju vitamíny, prolívám čajem a doufám, že to ze sebe vypláchnu, než ve 3 budu muset do práce 😪
@bagder I just realized that curl does not exclude localhost from proxied connections (while e.g. Firefox says 'Connections to localhost, 127.0.0.1/8, and ::1 are never proxied.'). Is that intentional? I can't imagine why one would try to access localhost via a proxy, but I bet you've seen more obscure usecases ;)
curl doesn't exclude anything from the proxy when you ask it to proxy, yes. That's always been so. The idea was always that you ask to proxy when you want to, otherwise you don't...
We deploy a system wide `HTTPS_PROXY` var, and in most cases that's exactly what we want -- all comms should go via that proxy. Just localhost was unexpected.
I need to reinstall Windows on my Surface tablet, and I wanted to see if I could use AI to get into the boot menu and change the boot order to try USB first. I have a Chat GPT subscription, so I used voice mode with the video camera on, gave it some information about what I was doing, and let it be my eyes for a while. I was delighted when the assistant actually knew how to get into the boot menu. It told me I could navigate the menu with my volume keys and select an option with the power key, much like navigating an Android phone’s recovery menu. I had to remind it I was totally blind a couple of times, but eventually, it helped me select USB as the first boot device and choose the “Exit and Restart” option. This was quite a few steps, and I was genuinely impressed at how easily I was able to fix this problem without interacting with a real person. I eagerly waited for the Windows installer to come up. It never did. So I used OCR on the screen, and discovered that not a single thing chat GPT claimed had happened … had actually happened. It didn’t hallucinate just one thing. It hallucinated an entire multi-step interaction with the firmware of my tablet. It basically experienced a break from reality for about two minutes and started describing what it thought should happen, with no regard for what was actually happening. Last week, the same app helped me learn the control panel of my heated mattress pad. It does work sometimes. But today, it led me on the wildest goose chase I’ve ever been on. I was actually trying to boot from the SD card, and as it turns out, that’s not even an option in the boot menu. But I made the mistake of teling Chat GPT exactly what I was trying to do, so it had all the material it needed to hallucinate a complex interaction convincingly. Never let yourself forget the all important “A” in “AI”. That intelligence is not artificial as in “synthetic”, it’s artificial as in “pretend”. No LLM has the slightest idea of what it’s doing or saying. The companies that create these models have the all-important task of trying to make their intelligence more convincing than every other company. That means they work most of the time, but the rest of the time, they will confidently lie. And that lie might be a missing digit, or it might be a whole entire interaction with a device. I called Aira and got it sorted. I actually had to touch the arrows on the touchscreen to rearrange the boot order. Yes, I had a keyboard connected. No, there is no documented way to rearrange boot devices on a surface using the keyboard. Yes, everythyng about this is moronic. But, it’s done, and once I get Windows on my USB device, I’m pretty sure it will boot because an actual human told me so. I can’t even begin to enumerate the possible clusterfucks that could arise from AI weaving such complex webs of lies. Do not use this shit for anything mission-critical. Ever. Even if it told the truth the last 99 times. Eventually, it will lie. When it does, you’ll have no idea.
Doesn't help you in the slightest but I've had really good luck with just the *free* version of Gemini Live. I'd A/B test it again Chat GPT with the same scenario out of pure interest.
@FreakyFwoof I did also figure out that, for whatever reason, I was using 4O instead of 5 at the time. Doesn't excuse the batshit crazy halucination, but I can at least say it wasn't the *latest* GPT model that did it. I will try Gemini. I've seen it lie but not quite as blatantly as GPT.
@FreakyFwoof I'm sorry to say this issue isn't, in my experience, the model, the whole thing is founded on sand. The only option, if you must use the models for this sort of thing, is to do things step by step, ask specific questions, and do so having understood the process from an external source. BTW, on a bit of a tangent, did you try the Be My Eyes microsoft system? It's free, as opposed to Aira. I've never used it for the surface, and haven't used it at all for quite a while, but don't see that they wouldn't be able to guide you through the UEFI. Did you try them and have issues?
@techsinger @FreakyFwoof I actually intended to try them first. microsoft laptop, Microsoft OS. But they were closed at the time. I'm up at an unusually early hour today. I need to install Windows on yet another old Surface so I will try calling them for that. I never have, and it seems like a missed opportunity.
@FreakyFwoof I would be interested in whether they can help. BTW, and again on a tangent, I haven't used a surface Pro for ever, but on my surface go, I was able to tell it to reboot into the advanced startup options and then boot from USB. I don't remember the exact steps but remember the whole thing because I thought I would need a human but didn't. I was surprised when ctrl+win+enter actually brought something up.
@techsinger If I had big big big big big powers, I would mandate that all companies adopt even ESpeak in the bios with a hotkey, so you could do whatever needed doing without help.
@techsinger @FreakyFwoof That should have worked, but for whatever reason it didn't. So, I tried to completely corrupt the boot manager. But the damn thing still insisted on trying to boot into Windows boot manager before anything else, and instead of falling back to the next device it just complained that nothing bootable was found. I'm gonna try advanced startup on my other Surface. I think maybe the docking station I used wasn't passing keyboard input through properly, and I don't have the official type cover for either model so I need to share one port between the install drive and the keyboard.
@FreakyFwoof I hate that sort of thing. It's the stupidity of devices which drives me up the wall. All I can suggest is the most basic plugs and cables, anything that calls itself a "docking station" is usually too smart for its own good, fails in some weird way, or otherwise makes you want to tear your hair out in handfuls.
@FreakyFwoof Yes. The chances of weird failures are less and they're cheaper. What a life! I understand, though, it's against the law to put more than one USB port on a device in Washington state. Life in prison for all devs and sellers, unfortunately.
That makes perfect sense. The big thing these models do is figure out the next word, or part of one, should be. They work based on the entire interaction you've had with them in the current conversation, using their training data to figure out the probability of the next thing based on all the current things. So if you're already talking about recovery screens, the probability loop will get more focused on the perfect and most satisfying recovery screen it can predict that you would think of as the best response, given everything that you've already talked about so far. Literally every word that either it produces, or you give it, will focus it on that ideal recovery scenariothat much more. A lot of people don't understand that, but that's what GPTs do.
@tcikoritys This is very well said. And yes. you can tell GPT to roleplay whatever you want, and it will do it convincingly. Sometimes, it will just go ahead and initiate a roleplay session without you asking first.
You know the cool part about that? For some of them, both their data and their prompts tell them to do that. can't use your web search tool? Fine, just tell them something, anything, just don't tell them you can't do that. I've seen both that in a prompt, and that in practice. Of course, when you call it out, the next predicted response is something like "Haha, you got me there! You're right, I can't do that right now. Let's try again, carefully this time." ... And then it does, getting it even more laughably wrong.
@tcikoritys It's such a perfect storm of convoluted tech that tries so hard to pretend to be conversational, when in reality it's literally making things up as it goes. You can't even ask how it came up with its nonsense answer because it doesn't know. You'll get a different answer every time. You can feed the same context to a diffforent model and it will be forced to accept the fact that the conversation happened the way you claim it did. In a vacuum they're a lot of fun to play with. They are surprisingly helpful sometimes. In the real world, people don't understand anything about how they work, companies don't facilitate that understanding because it doesn't benefit them, and the world conflates fancy autocomplete with the robot uprising.
Ahaha, totally reminds me of me trying to get secure boot disabled with GPT - it confidently kept saying I turned it off, that I'm in the advanced tab and just press arrow down a few times to find it. Of course, I was in the boot tab, and I later resorted to asking it more limited questions because the moment I would give away what I wanted to do, it would get way worse. No asking it, "Look for somethhing with the label secure boot and tell me how many arrow downs it is." Had a similar experience trying to get it to read the MAC address off a router, but at least there I could use something more local like Seeing AI and get an answer that I knew was as accurate as the ability I have to keep a good angle for it to read it. I think that's the other problem with GPT or other AI: It doesn't inform you like a human (always) that you need to move or tilt your phone. I've had it help me with that a few times, but unless you specifically ask it whether your phone needs adjusting, it would rather make up the thing than tell you.
I wonder if we could finetune a small VLLM specifically to describe photos and read text as accurately as possible without being able to do much else? I think the issue is these models are designed to be generalists, so they're not really good at any one particular thing.
Maybe I've said this before...Maybe I haven't? But I feel like Apple should rename VoiceOver. So, so often, I encounter people who hear "VoiceOver" and think they can control their devices entirely by voice. It says "Voice", after all! And we spent a long, long time discussing and looking at how it works and how you can't talk to it, and people seem disappointed in the end, especially in the beginning before they pick it up. Maybe that's just the process of learning overall, but still.
It's a cool name, I like voiceover. It's original. Not everything in this world needs a boring name, like talkback, or simply screen reader, or even apple screen reader, or whatever varians thereof. NVDA and voiceover are super named. Love the names. Jaws is a shark, nobody cares about his overpriced butt, although he was a nice shark years ago. Voiceover is cool, if blind relatives need use with voice, voice control is the right term. It even says voice control. Voiceover is like understandable it's not controlled by your voice, if people know the meaning of voiceover. Originality and creativity is gone. Makes me sad everyone complains instead of researching the difference between voiceover, and voice control. Also, it says what it does when you click on it. In both instances, its purpose is stated in the settings. People have gotten to learn how to read, before writing nonsense. And voice control is the answer to everyone who wants to poke phones with their voice.
My further supporting argument, voiceover means voice over something. It can be the phone. People are too lazy. Apple screen reader is a dumb thing as a name, and comes out of laziness because people don't wanna read what it says on the label. Or I am just old, grumpy, and I like apple. Maybe those. Cackles.
@tardis I agree with all of this. I like the name, too, but it seems to confuse people who are new to this technology. It sure does explain a lot right on the settings screen, but people don't read that shit, sadly. Just observations I've seen.
Not sure what's confusing. voiceover means acting, having voice over something, not voice control. People need dictionaries and access to a large infodump.
I hate AI customer service support. You need to fix or solve something, call the customer service and AI agent tells you everithing what you already did. Finally you get to the real person but it takes longer time. To be clear, I use CHATGPT, Auphonic and other AI tools but for customer support it doesn't work.
NVDA 64-Bit Development. Important add-ons which function flawlessly as of this writing: Eloquence, Enhanced Find Dialog Modification, ObjPad, Say control type before label, Say product name & version, SentenceNav, Speech history review and copying, SysTrayList, TextNav, Thunderbird+G5, UnigramPlus, Virtual Review, WhatsAppPlus. Important add-ons which don't work or malfunction as of this writing: 1. ColumnsReview: can be loaded with the Compatibility flag, but doesn't function at all. 2. MSWord Accessibility Script: can be loaded, but doesn't work. In fact, if it's active, MS Word cannot be used with NVDA. 3. BrowserNav: can be loaded, but most features either don't work or malfunction. For instance, it can't find its bookmarks set on various websites. 4. Tony's Enhancements: issues an error message and cannot be used. @NVAccess
Hearing the stories of peoples iPhone battery usage, I'm convinced some bug in my iOs setup is eating up battely. Took my phone off of a charger at 4 this morning. A.few minutes of social media, 2 boarding pass scans on full brightness, a podcast download over wifi and 2 hours of podcast playback with about 20 minutes of taxy ride took it down to 40 percent by 10 in the morning. And its a pro max, not an air. I guess starting my setup from scratch is in my near future.
@vick21 I tend do reset and start over every couple of years for this reason. Every time I do notice longer times between charges. I'm sure its particular apps that cause this issue over time but finding the right one would take longer than a reset, plus you get newer defaults by doing the reset.
@vick21 I would say 10 years is well overdue a refresh. There are a lot of defaults that got changed over the years as well which you will get by refreshing.
@andrew @vick21 I am not running betas right now, but your approach seems to confirm my suspicion. MY last set up from scratch was over 10 years ago, and I'm quite confident, that there are underlying bugs and undetected issues draining the battery.
So here's the scoop: LCUs (Latest Cumulative Updates) are language-agnostic — but their payloads aren’t. wow. The .msu itself is universal, but the moment DISM starts applying it, it looks for language satellite components (MUI files, language resources, etc.) under WinSxS. If the base image has en-US only, the update touches a small number of components and sails through. If the image has 10+ languages, every single one of those components must have a matching resource payload — or the update installer logs a “missing payload” error and bails mid-way. Occasionally, a later LCU doesn’t have the full set of MUI payloads and expects them to already exist. If they don’t, it throws 0x800f0823 or a similar “pending component” error. On a single-language image, this never happens because there’s nothing to mismatch. For language packs, downloading them as offline installers is limited: Windows Update Catalog rarely has full LP CABs; it’s mostly LCUs/SSUs. Visual Studio Subscriptions or the volume licensing center have them. So no, I cannot grab the latest language pack files easily.
Soon your data like your resumes and profile information will be used to train LinkedIn, Microsoft, and Co.'s artificial intelligence. And no, you didn’t opt-in. Take action and stop this now.
I’m soooooo tired of every online website changing Terms of Service to the infliction of AI theft — and the need to actively opt out - if that’s even an option.
- As if they haven't trained AI on it already, this is just formalising it because lawyers got involved - As if they haven't sold the data to third parties who have trained AI on it already
Definitely delete your account, but the horse has surely bolted already for almost all data online whether they tell you about it or not.
darn yup. For custom installs, the new multilingual ISO is trash. For upgrades too maybe. all of it. I was able to get passed 88% on my original image. Things are installing like normal. Sadly you lose the in-place Win11 upgrade, but we might have to live with this until someone smarter than I comes out with a proper ISO. Sorry y'all, I tried. After a few tested it though and I tried to custom install, I'm even strongly considering removing the links to not mis-lead people with a broken ISO. At least the original one, while English only, can install just fine when you choose English US. This setup if booting from the media also gives you "upgrade" and unlocks it, so it may be that you can upgrade from Win11 that way, but I haven't tried without a proper Win11 VM (might do this on the weekend to do more testing though.) Either way, it's a bust. Will probably try to super-focus in on work and not think of this terrible disaster.
I finally did it. I've practically maxxed out my computer over the years. I can't add anything there, all slots are taken, all drive positions are filled, I can't put a better CPU there, the only viable upgrade now is to go from 64 GB to 128 GB RAM and I don't need that.
I guess my next motherboard one day will have to be an E-ATX form factor to have more options and an over the top chipset to handle it all.
I am exactly in the same situation. I maxed my PC that I use as my all in one server few years ago and now I just looked around what could be my new motherboard. And I was surprised that all usable (and the count isn't large) cards are E-ATX.
Zajímavé, taky jsem si včera koupil boty, ale na chůzi. No některé byly takto odlehčené a odpružené, ale vzhledem k mé hmotnosti, jsem se na nich cítil jako by byly balanční 🤣 Se vůbec pro mě nedalo použít 🙈
@chick_enheart Problém je jak mizí úspěšné modely a nové stojí za prd. Co se týče nízkých odlehčených trekovek GTX (na cestu do práce), od Adidas přes Solomon, Merrell jsem musel letos jít do Saleva z Německa. Prostě odlehčené modely s širší botou + GTX - vždy první vyprodané - po pár sezonách přestanou vyrábět 🤷♂️
)
Andre Louis
in reply to Kevin R Jones • • •Kevin R Jones
in reply to Andre Louis • • •