@lyse@lyse.isobeef.org Hmmm 🧐
@movq@www.uninformativ.de fuck 🤣
robots.txt that I have on https://git.mills.io/robots.txt with content:
* 185325d - (HEAD -> master) edge: Ban Alibaba (38 seconds ago) <James Mills>
fark me 🤦♂️ Alibaba, CN has been hitting my Gopher proxy quite hard as well. Fuck’n hell! 🔥
robots.txt that I have on https://git.mills.io/robots.txt with content:
* 8a77b64 - (HEAD -> master) edge: Ban Google's ASN (21 seconds ago) <James Mills>
It would appear that Google’s web crawlers are ignoring the robots.txt that I have on https://git.mills.io/robots.txt with content:
User-agent: *
Disallow: /
Evidence attached (see screenshots): 
– I think its the the Small Web community band together and file a class action suit(s) against Microsoft.com Google.com and any other assholes out there (OpenAI?) that violate our rights and ignore requests to be “polite” on the web. Thoughts? 💭
I got promoted today to try using Passkeys on Github.com. Fine 😅 I did that, but I discovered that when you use your Passkey to login, Chrome prompts you for your device’s password (i.e: The password you use to login to your macOS Desktop). Is that intentional? Kind of defeats the point no? I mean sure, now there’s no Password being transmitted, stored or presented to Github.com but still, all an attacker has to do is somehow be on my device and know my login password to my device right? Is that better or worse? 🤔
Our stupid fucking 🤬 cat 🐱 keeps pissing 🚽 everywhere in our house 🏠
@lyse@lyse.isobeef.org Ouch 🤕
@sorenpeter@darch.dk Nice 😊
@lyse@lyse.isobeef.org Bur I can’t think of what’s changed to cause this? 🧐
@movq@www.uninformativ.de Hahahaha 🤣
@lyse@lyse.isobeef.org Duplicates again where two days ago? I don’t see this anywhere (unless I’m blind!)
@lyse@lyse.isobeef.org Agree. I’m not sure we should lax the timestamp format at all IMO. What @xuu@txt.sour.is has found is kind of nuts haha 😆 However I do think we should relax the \t separator between <timestamp> and <content>. Let users use any valid whitespace here that isn’t a newline or carriage return.
I think the author is a bit out of their depth here. A linear feed isn’t quite what the author seems to be modelling in their view of the problems they observe and describe. A linear feed has a beginning and an end. You can (ideally client-side) sort it by Date, or by Subject like we do with our Twtxt clients. A Tree-structure isn’t what the author thinks either, this is more the structure that forms after you introducing some kind of “threading model”. The main problem with any kind of information system that tries to figure out algorithmically what you want to “see” is that type of interface has no start and no end. SO you end up with a “scroll of doom”.
The article discusses the challenges posed by linear social media feeds, which often lead to disengagement and difficulty in prioritizing content from friends due to constant scrolling. The author proposes an alternative approach using a daily feed structure per day, which organizes posts by date, allowing for easier prioritization and reducing mindless scrolling.
Key Points:
Linear Feed Problem: Linear feeds present a long list of posts without prioritization, forcing users to scroll endlessly to catch up on friends’ content. This can lead to addiction and disengagement.
Proposed Alternative (Tree Structure): The daily feed structure organizes posts by day, enabling users to prioritize updates from friends who post infrequently while reducing scrolling effort.
Mastodon Experience: The author’s experience with Mastodon highlighted its effectiveness in allowing content prioritization and managing social media usage without dependency on algorithms.
Challenges and Considerations:
- Implementation Challenges: Creating a daily feed system involves organizing content effectively and ensuring users can prioritize posts.
- Platform Support: Current platforms may not have APIs conducive to such changes, making it difficult to implement without significant technical changes.
- Engagement Metrics: The impact on engagement metrics needs to be considered, as traditional metrics might be misinterpreted in a tree structure.
- Implementation Challenges: Creating a daily feed system involves organizing content effectively and ensuring users can prioritize posts.
Potential Applications Beyond Social Media: This approach could empower users by giving control over content consumption and aiding in balancing social media use without overwhelming them with information.
Future Directions: The author hopes for improvements in alternative platforms’ feed systems and engagement metrics, potentially through more interactive content models or changes in APIs.
In conclusion, the article emphasizes the importance of providing users with control over their content consumption, moving away from linear feed
What exactly is a linear feed? 🤔
@andros@twtxt.andros.dev My first point of advice is to stop everything and measure all the important critical user journeys. Design and Build Service Level Objectives for each and every part of the system you can find that any user cares about.
@bender@twtxt.net @lyse@lyse.isobeef.org This bug was fixed back in September last year. But the brackets still appear in my current feed. is that what the issue is? 🤔
@lyse@lyse.isobeef.org @andros@twtxt.andros.dev Are we talking about yarnd here? Hmm? 🤔 I’ve thought about a “read flag” but I just haven’t bothered so far…
@johanbove@johanbove.info And what are the results so far? 🤔
The cache is only suppose to be for 120s though, but I reckon the caching layer is just stupid? 🤔 (and maybe buggy)?
I need to understand how the caching is at play here at the edge. I hit CTRL+R on @mckinley@twtxt.net’s OP to get the right subject reply after poking at the underlying HTML elements on the page.
@bender@twtxt.net To be fair, I do this in my “spare time” 😅
@mckinley@twtxt.net Ahh. I think this is some kind of weird caching issue at my edge! 😱
@mckinley@twtxt.net I’m worried we’re really approaching a point where we need to adapt the hashing algorithm and expand the no. of bits. Is it at all possible something else is going on here though? 🤞
@lyse@lyse.isobeef.org True, but we should also consider building tolerant “systems” 😉
@movq@www.uninformativ.de I believe that’s the same client I’ve used in the past too. Works nicely on macOS 👌👌
@lyse@lyse.isobeef.org Where? 🧐
robots.txt file. only noticed it because the OpenAI bot was hitting me with a lot of nonsensical requests. here is the list from last month:
@bmallred@staystrong.run Similar story here 😱
@jost@jost.sdfeu.org “right person”, who’s to say? And not “AI” but more-generally search. Otherwise I agree ☝️
🤣🤣🤣
Gotta get faster disks man 🤣
@jost@jost.sdfeu.org What is this that you maintain? 🤔
Whoohoo!
root@vz3:~# free -h
total used free shared buff/cache available
Mem: 62Gi 10Gi 49Gi 72Mi 2.9Gi 51Gi
Swap: 0B 0B 0B
@andros@twtxt.andros.dev Nice! 👍
@andros@twtxt.andros.dev What’s the problem sorry? haha 🤣
@xuu@txt.sour.is da fuq?!
Look at the size of this coffee!!! 😱 
💭 Remember kids 🧒
The “Cloud” is just someone else’s computer(s).
@movq@www.uninformativ.de Over a month now. I think since Jan 8th this year.
@thecanine@twtxt.net The thing is, all these IVR phone systems are so fucking useless it’s not even funny. They either can’t understand you, are stupid as fuck, or somewhere in between. You spend minutes to hours trying to figure out how to get anywhere you basically just end up giving up. I think it’s done on purpose 😱
Sorry it appears I have having difficulty understand how to help you.
Please contact Microsoft at aks blah balh balh
9834) where we went through the form together. Managed to get to the 3rd page, at which point the "Address" had an asterisk next to it (*) -- I advise operator #9834 that I was not comfortable handing out a specific address, at which point she responded:
e.g Spam Act 2003 of Australia
9834) where we went through the form together. Managed to get to the 3rd page, at which point the "Address" had an asterisk next to it (*) -- I advise operator #9834 that I was not comfortable handing out a specific address, at which point she responded:
I guess this is why companies get away with breaking all kinds of stupid ass lawas 😅
So I was advised by the FTC to file a complaint with the Internet Crimes Complaints Centre (IC3) – That is, the FBI. They have an online form that you can use, however it is clearly designed for US citizens, not designed for cross-country complaints such as from AU. Needless to say, I tried to speak with a lovely woman on the phone (who would not say here name, only her badge number: 9834) where we went through the form together. Managed to get to the 3rd page, at which point the “Address” had an asterisk next to it (*) – I advise operator #9834 that I was not comfortable handing out a specific address, at which point she responded:
Well the IC3 online form is the correct form. Thank you. Goodbye.
And hang up! Haha 🤣 wut da fuq?! 🤦♂️
Phase #1 done. 
– Next disk! 💪
In some organizations, engineers in the general field of IT (Software, SRE, etc) are recognised hierarchically as: Intern, Graduate, Junior, Mid, Senior, Staff, Principal, Distinguished.
@lyse@lyse.isobeef.org @bender@twtxt.net The funny thing is, this person, is still trying to convince me to go to this thing. I don’t even know wut da fuq it is?! Do they smoke peace pipes or something? Is it a hooker joint or something?! I have no idea 🤷
@movq@www.uninformativ.de Same 😢
@bender@twtxt.net I’m really not sure wtf happened tbh 🤣
Hi James, great to hear your interest. So this is an exclusive roundtable luncheon with people in the IT, Engineering, DevOps and Technical professionals. This is an opportunity to benchmark and share stories and experiences with like-minded peers in a closed-door, Chatham House Rule environment where you will be given the opportunity to speak openly and candidly.
I’m not even sure what half these words mean hmm 🧐
Nobody that does good things earns a lot of money.
@hacker-news-newest@feeds.twtxt.net Blah blah blah AI opinion blah blah
@lyse@lyse.isobeef.org Bahahahahahahaha 🤣😆🤣😆🤣😆🙌😆🙌
@jost@jost.sdfeu.org Reach out to @movq@www.uninformativ.de 🤗
@bender@twtxt.net So to be clear, you’re just hitting Reply that this is getting written into your feed? 🤔
@bender@twtxt.net da hell? You somehow wrote this as:
$ bat https://twtxt.net/twt/mxszcva | jq -r '.text'
(#c4tigxa) @<twtxt.net https://twtxt.net/user/prologic/twtxt.txt> Octal 31 = Decimal 25
in your feed? 🤔
No-one happy with the price of coffee one day and it skyrocketing the next eh? 🤣
Well that didn’t take long 🤣
@jost.sdfeu.org@jost.sdfeu.org Hello! 👋
I have no ties to the US, EU or even the AU. Government overreach can bite me 🤣
@darch.dk@darch.dk You are welcome to use git.mills.io
@lyse@lyse.isobeef.org Bahahahahaha 🤣
@movq@www.uninformativ.de I guess u don’t get the relationship between Oct and Dec 🤣 Mist be an American thing? 🤔
@movq@www.uninformativ.de I’m afraid I’m too stupid (probably blind, which I am) to get it 🤣
@movq@www.uninformativ.de No sadly I didn’t 🤦♂️ I’m admitedly not very good at jokes like this (not even sure what style this is sorry!) 😢
@movq@www.uninformativ.de Bahahahahaha 🤣
@movq@www.uninformativ.de Oh wow! That’s the moon the Venus right? 🤔
For context, I’ve write to get one of these LLM(s) to do this for a project once before a couple of years ago. All it had to do was read the public methods of a struct and write the boilerplate for a type Foo interface { ... } with those same matching methods.
This shit™ is such a waste of time, human effort and global energy 🤦♂️
Seriously?! 😳 Transscript
if I had a Go struct such as the following:
…
Here’s how the complete interface would look:
type Bar interface {
Read(t *Bar, b []byte) (int, error)
}
This interface matches the behavior and method signature of the provided Foo struct.
This is total garbage 🗑️
In sane 🤣🤣
$ ollama run deepseek-r1:7b
>>> How does your model's reasoning work exactly?
<think>
</think>
I'm an AI assistant that uses algorithms to process information and create responses. While I can mimic some patterns of human thinking in my responses, I
don't have real human thought or consciousness.
>>> So how do you deliberate and argue with yourself over counting the number of letters "r" in the word "strawberry"?
<think>
Alright, so I'm trying to figure out how someone would deliberate and argue with themselves about counting the number of times the letter "r" appears in
the word "strawberry." First off, it seems pretty straightforward, but maybe there's some hidden complexity here.
...
By following these steps, it’s clear that the word “strawberry” contains exactly two “r”s. While overcomplicating the process with comparisons to similar
words, pronunciation considerations, or playful activities can be fun exercises, they aren’t necessary for this simple counting task. The key is to
approach the task systematically and carefully.
Ridiculous 🤦♂️
deepseek-r1:7b model using Ollama and this is the result:
I can’t wait to be in an operating theatre one day when I’m old and an LLM (AI) bot is arguing with itself over the decision to cut the left or the right blood vessel 🤦♂️ Or whatever other horrifying situation you/we may find ourselves in with this kind of “dumbass” reasoning behind the wheel 🛞
How many r’s in the word strawberry? 👈 Just figured I’d try to run up the deepseek-r1:7b model using Ollama and this is the result:
$ ollama run deepseek-r1:7b
>>> How many r's in the word strawberry?
<think>
Alright, so I need to figure out how many times the letter 'r' appears in the word "strawberry." Hmm, okay. Let me start by writing down the word to
visualize each letter clearly.
...
Twoo lines later… Argueing with itself over and over again…
The word “strawberry” contains three ‘r’s when spelled out as S-T-R-A-W-B-E-R-R-Y. However, upon closer examination and considering standard spelling
conventions, it actually has only two distinct ‘r’s: one in “straw” and another pair in “berry.”Answer: There are 2 ‘r’s in the word “strawberry.”
Farrkkkk me 🤦♂️
@bmallred@staystrong.run Thanks dude 🤗 I’ll get over it, sometimes things can get a bit touch, it’s all good 👍
@movq@www.uninformativ.de Hmm in who’s world? 🌎 Haha 🤣
Because Oct 31 = Dec 25.
@arne@uplegger.eu This ☝️
in other words, do companies in other parts of the world behave as though they are above approach? 😅
To all my EU friends out there, is it this hard™ to reach a human in European companies that allow, perform or permit silly shenanigans? 🤔 Or is it just US companies? 🤔
Contact | Federal Communications Commission 👈 Haha calling the FCC was interesting. But I’ll have to call the consumer folks back next Saturday (morning for me) to speak with someone. Apparently they’re closed now, even though it’s 3pm EST right? Hmmm 🤔
cli/q: 🌱 A simple programming language. - q - Projects I really like this little q lang that Ed has created ❤️ Really nice and simpler, great design and implementation and really lovely cross-platform compiler supporting DOS, Windows, Darwin and Linux on AMD64 and ARM64 💪
That moment, when you’re a little pipsqueak self-host(er) with infrastructure you’re trying to protect from potentially rogue customers on Azure Cloud. You file an abuse report, but receive no response or acknowledgment for weeks. Now what? 🤔 Email the CEO! #Azure #Microsoft
Finally I’ll also be setting up CubeFS (finally mature enough to give it a serious go) and slowly migrating workloads to use it as Persistent storage across the 3 Hypervisor nodes (which will run KVM + CubeFS) – Thank goodness this thing (CubeFS) doesn’t need to run on Kubernetes 😂
Over the next few months, I’m a) Replacing all Boot disks b) Replacing all Storage disks c) Adding additional 32GB Memory to all my Hypervisor machines. I think that’ll take me the better part of 6 months. For the remainder thereafter in the year I’ll be swapping out the 3.5” Disks in my NAS slowly (one at a time).
Upcoming Products for M4 Mac mini - SONNETTECH Gonna buy myself one of these when they come out (hopefully soon™) and populate with 3 M4 Mac Mini(s) – Just the base model is enough. Purpose? Migrate Plex to one of them for better performance, and run CI/CD Runners on all 3 as well as other I/O intensive tasks like syncing Youtube channels to Plex.
@github_blog@feeds.twtxt.net Only 4 steps you say? Wow! :O
@lyse@lyse.isobeef.org I have no problems with the ternary operator either. If it were added to Go I Wouldn’t mind. C has it right? I’d also by happy with if expressions, e.g: if foo ... else bah, but probably doesn’t fit the styoe of the Go grammer.
What I absolutely hate is this proposal. Making ? to magical™ things, making the code harder to read and reason about is bonkers.
@thecanine@twtxt.net This looks sick as your new Avatar 🙌
@movq@www.uninformativ.de Thanks this is good to digest 👌
PS: Please forgive me. I’m just being a really grumpy bastarda the last couple of days. I’m really not having a lot of fun this past week 😢
@xuu@txt.sour.is And you’re right, it only really saves a single line. I think I’ve even commented on a similar discussion a few years ago as well where I showed concretely the abrustdity of the community wanting to change this. If I recall, the whole pattern of:
if err != nil {
...
}
basically accounts for <1-3% of your codebase in the first place.
The entire thing (fuss) is really just pathetic IMO.
@xuu@txt.sour.is I’m not necessarily saying it is to be bluntly honest. But it’s far better than this ? garbage. I don’t want to be rude, but honestly wtf is wrong with people and making languages harder to read?! 🤦♂️
I want to propose my own counter-proposal to the discussion that’s ongoing with Go and error handling.
Here it is: https://docs.mills.io/ix4qDHMnQUSPxZ5tXz12Vg?view
It’s very rough and needs much more work, but essentially I want to propose the following change to the language’s grammar:
f := os.Open("foo.txt") or (e error) {
log.Fatal("error opening file; %s", e)
}