• 40 Posts
Joined 1Y ago
Cake day: Dec 19, 2021


I know that patent trolls are a thing, but I was under the impression that these patent trolls would still have to produce lots of specific but still broad patents. For example, they will file a patent for a whole class of compounds for treating a specific group of diseases without having the data to back it up.

But in this case it is even worse!

They patented a trivial idea - “send voice commands from phone to computer” - without providing any novel practical implementations of the idea. How is this acceptable?

patent/US9794348 Using voice commands from a mobile device to remotely access and control a computer


A method of using voice commands from a mobile device to remotely access and control a computer. The method includes receiving audio data from the mobile device at the computer. The audio data is decoded into a command. A software program that the command was provided for is determined. At least one process is executed at the computer in response to the command. Output data is generated at the computer in response to executing at least one process at the computer. The output data is transmitted to the mobile device.

The MAX_TITLE_LENGTH is applied to posts that are fetched from other instances using activitypub. The posts that are posted within an instance do not pass through that processing step.

Yeah, no worries, I have been looking at how the Ohio situation is unfolding too

I think you are referring to a different incident - the East Palestine train derailment in Ohio, right? I agree, it is baffling.

Well, of course, what other response can anyone expect?

"White house says blog post is factual, admits to blowing up Nord Stream - 'And we would have gotten away with it too, if it weren’t for that meddling journalist 😠 ', the White House representative added "

Yes, I agree with your assessment. I think that from the point of view of geo-politics, countries tend to make self-serving choices. Human rights are usually not a priority. There are human rights abuses all over the world, but if addressing them does not serve a political purpose, they tend not to appear in the spotlight. As soon as an excuse is needed one can of course summon these abuses and blast them through the media.

I am not saying that I agree with this approach or that this is the way I think thinks should be. But I think that this is the answer to why the EU does not impose sanctions to the US. Similarly, the US lets the Saudis get away with a lot of human rights abuses to avoid souring relations. And the Israel <-> Palestine conflict, why is the world so chill about that?

Also, the sanctions did impact the EU economy and the leaders have had to pay a price. They have gone as far as they have in large part because the US has pressured them - the EU possibly would not have several of the more drastic measures if it was up to them, to protect their economy.

Because the US is a closer ally and the EU was forced to pick a side. The EU did not move spontaneously and alone, it moved as part of coordinated action. Furthermore, Russia launched a direct attack into European territory, making it a much more direct military threat.

What you are proposing here is different. You are asking the EU to spontaneously and unilaterally decide to go after its most powerful ally because of the human rights violations that they are committing somewhere else.

Sounds to me like that would be a very expensive choice that would severely affect the quality of life of EU voters. Since those in a position to impose sanctions are those who play the game of getting elected, I don’t think that it is in their best interest to tank the economy.

Yup, this was fixed in version 0.0.24! No worries.

What version do you have? I think you have one of the older versions, you can find the latest one here:

I think that the newest update changed something such that the the instances that updated to 17.0 are no longer being recognized by the crawler. So the crawler hasn’t been updated to deal with the changes.

I am not convinced that the people who have spear-headed several of the crypto projects are being dishonest to make money. So, then, who is the scammer? A pump-and-dump scheme is a scam, but that does not make the underlying asset a scam. The market price of growth stocks is extremely inflated, but that does not mean they are a scam. These are speculative assets that are expected to appreciate due to the self-serving feedback loop of the market.

From an investment perspective, crypto can be seen as a highly speculative volatile asset with a value that is largely decoupled from intrinsic value. That does not make it a scam. Someone can scam someone else by making guarantees about the future - but it is entirely possible to do this with any other asset.

But, more importantly, crypto is practically useful. What is a better way of paying for drugs online than using monero?

Great job figuring that out!

At some point it would be fun to work on some new guides anyway :-)

Oh… Yes, I can. I had not encountered before a tor-only clearnet url. That’s cool.

Url appears to be broken at the moment

“There is no law against prisoner organ donation,” he said

the bill will “restore bodily autonomy to incarcerated folks by providing opportunity to donate organs and bone marrow”

🤔 🤔 🤔 🤔

I think that if you use the new docker-compile file, the one that includes the NGINX block, then you might not need to go through the step of the nginx configuration.

I have not tested this myself though, but I can test it when I make the updated tutorial.

I can do that, but I can’t right now, I am very busy writing for my thesis unfortunately :/

If I manage to write enough tonight I might be able to do it tomorrow.

Oh no! It is possible to save your SSL certificates by simply copying the files into your computer, then you don’t need to run the certbot again.

Try removing the lemmy.conf file from /etc/nginx/sites-enabled,

And then run:

sudo wget /etc/nginx/sites-enabled/uebishe.conf

sudo systemctl restart nginx

Or, I have made an nginx config file that will work for you

If you already created an nginx configuration file, move it somewhere else. Then, get the new configuration file to the nginx folder, and then restart the nginx service by running these two commands:

sudo wget /etc/nginx/sites-enabled/uebishe.conf

sudo systemctl restart nginx

Ah! So you skipped this step:

Now i skip old step about “wget -O lemmy.conf”

You need to have the lemm.conf file under /etc/nginx/sites-enabled, this is the file that tells nginx how to manage the connection. Don’t skip that step! Get that file and modify it by adding your domain name

Sorry, I did not realize that these files had not yet been updated with the changes necessary for version 0.17.0, which are described here:

I could not find an example of the files configured correctly in github, so I have uploaded some example files. You get them this way:


Then modify the variables inside {}, like the domain name and the database password.

Remove the following blocks from the docker-compose file:

  # communication to web and clients
  # communication between lemmy services
    driver: bridge
    internal: true
    image: nginx:1-alpine
      - lemmyinternal
      - lemmyexternalproxy
      # only ports facing any connection from outside
      - 80:80 
      - 443:443
      - ./nginx.conf:/etc/nginx/nginx.conf:ro
      # setup your certbot and letsencrypt config 
      - ./certbot:/var/www/certbot
      - ./letsencrypt:/etc/letsencrypt/live
    restart: always
      - pictrs
      - lemmy-ui

And try again. Does it work?

Do you want to install a fresh instance, or do you want to upgrade an instance?

If you are installing fresh, most of that tutorial is still going to work out. But you now need to pull the config file this way:

wget lemmy.hjson

If you install the more recent docker version, you don’t have to install docker-compose the same way. Follow the instructions from the website, and use ‘docker compose’ instead of ‘docker-compose’.

The nginx configuration does not need to change.

What problems are you getting?

An issue was opened for that three days ago:

I don’t think it is possible through the UI at the moment.

Oh, that looks like an error in your lemmy.hjson settings file. If you share that one too with the passwords redacted I can check.

If you go to your lemmy directory and type docker-compose logs -f lemmy, what does it say?

Can you share your docker-compose.yml file, removing the database password and any other private information?

Also, did you change the variables in the Lemmy UI environment?

(Edit to specify, in the docker-compose.yml file the variables are now different)

The lemmy-ui environment variables have changed, and should now look like:

    - LEMMY_UI_HTTPS=true

There are also other instructions that you need to follow. Are you following the release note instructions?



docker compose logs -f lemmy

(or docker-compose) to see the logs.

Is the last message saying that the database migration is going to be run? If so, you need to wait. It can take a few hours.

I hope this gets stopped.

Question for anyone who knows the details - does this policy have something to say about self-hosted chat services such as a personal XMPP and matrix servers? Or would this policy apply specifically to the big guys like WhatsApp, Signal, and gmail?

Me: If you help me write an academic paper, should I include you in the author’s list?

ChatGPT: As a language model developed by OpenAI, I am not a human author and cannot take credit for work produced by someone else. You should not include me in the list of authors for any academic paper that you write. The authorship of a paper should reflect the individuals who have made substantial contributions to the research and the writing of the paper, and using language models like me should be acknowledged in the acknowledgments section or as a tool used in the research.

ChatGPT agrees with me ;P

Added bonus: I asked ChatGPT whether they are okay with me referring to them as “it”, and they recommended using the gender neutral form they/them instead:

You can refer to me as “it” if you prefer, as I am an AI language model and do not have a gender. However, many people prefer to use gender-neutral language and use “they” instead of “it” when referring to non-binary individuals or entities. Either way is acceptable, but using “they” is more in line with current language conventions for referring to non-binary entities in a respectful manner.

It is a big debate now. My current view is that these are tools that we use, and they should be described in the methods section.

As we increase our reliance on these kinds of tools, it doesn’t seem practical or useful to me to include the name of the 17 AI tools that we will use to write a paper in the authors list.

But it is not a very developed view. I can be persuaded to think that AI tools deserve authorship! I would ask ChatGPT what it thinks about it but it is unavailable at the moment.

Shamelessly stolen from [reddit/r/piracy](

It shows that the community has been deleted by the creator, and the creator’s account has also been deleted.

You can see this by accessing the community directly via the URL from a federating instance, for example:

I am not sure about whether being able to access this information this way is a feature or a bug.

Yeah, it can be quite useful!

The discussion linked is not addressing whether it is suitable or not to use ChatGPT as a tool, but rather addresses whether ChatGPT should be given authorship.

It references these papers where “ChatGPT” was included in the author list:

Bookmark shortcuts
I just figured this out and it is so convenient. I will show by example. I am using Firefox, but other browsers might work the same way. (1) Go to (2) Ctrl+D, then enter to save as a bookmark (3) Ctrl+shift+O to open the bookmark settings (4) Select the lemmy bookmark that you just created (5) At the bottom of the menu, you can set a Keyword. Write the letter "l". (6) Close the menu You can now type "Ctrl+L" to jump to the navigation bar, then type "l", enter. You can create single or few-letter keywords for the sites that you visit often. Bonus related tip: In the browser preferences, under "search", you can define a keword for specific search engines. For example, if you set wikipedia to "@w", you can then search directly from the navigation bar by typing "@w whatever"

Do you heat the pan first, then add oil? Or put the oil in and heat up with the pan?
I personally heat up the pan first, then put the oil in and after it's heated up add the ingredients. I go with the line of reasoning that doing it this way gives the oil less time to burn, thinking that if you do it the other way, by the time the pan and oil has heated up, the oil could already be starting to burn. I've never experimented, but I think this is more of an issue with electric stoves since you can modulate the heat more quickly with gas, ie turn it off if the oil's starting to smoke. [Asked by Ward - Reinstate Monica to cooking/stackexchange](

How can I chop onions without crying?
Onions are an excellent addition to many dishes, but cutting them can be frustrating when they make you "cry" all the time. Does anyone know any tips or tricks to help minimise the tears when chopping onions? [Asked by Caleb to cooking/stackexchange](

If hand sanitizer kills 99.99% of germs, then won’t the surviving 0.01% make hand sanitizer resistant strains?
[Asked by /u/ClF3FTW to /r/AskScience](

Why do we have to “fall” asleep? Why can’t we just decide to be asleep?
[Asked by /u/SixthGrader to /r/AskScience](

I have created the community [/c/answered](/c/ If you run into an interesting question online for which a good answer was provided, this is the place to share it! The point of this community is to aggregate interesting questions that have been answered from all over the internet (like stackexchange, reddit, quora, etc). You can copy the questions into the post body and the best answer(s) into the comments. Other users may then address the answers directly by replying to the comments, or add their own answer to the question as a new independent comment. As of now, there is no constraint on the topics. Any answered question is fine! I have already made an example post, but I don't know how to create a relative hyperlink that would send you to the post in your instance. So just check it out in the community ;)

Why don’t metals bond when touched together?
It is my understanding that metals are a crystal lattice of ions, held together by delocalized electrons, which move freely through the lattice (and conduct electricity, heat, etc.). If two pieces of the same metal are touched together, why don't they bond? It seems to me the delocalized electrons would move from one metal to the other, and extend the bond, holding the two pieces together. If the electrons don't move freely from one piece to the other, why would this not happen when a current is applied (through the two pieces)? Asked by jcw in [physics.stackexchange](

How do you feel about this personal e-mail format?
Let's say as an example that my full name is Robert Frankel If I purchase the domain `` and use that to create my custom e-mail: `` And then use this as my personal e-mail in a professional setting (resume, contact e-mail, business card, etc..). Do you think this would be perceived as unprofessional and silly in a negative way, and even confusing to some? Or do you think most people would understand it right away and think of it in a net positive way - perhaps a bit silly, but memorable. Don't worry about whether the e-mail will be filtered as spam, I am curious about how it is perceived. I am asking because I have a domain like this that I use for private personal stuff, and I wonder whether it would be a good or bad idea to use that e-mail when applying for jobs, as a small extra detail to stand out 😁

I finally got around to setting up the e-mails correctly for my instance ( and noticed that the e-mail deliver-ability is awful, as I get heavily penalized by spam filters for having an .xyz domain. I would not be surprised if search engines penalize this too. It is not a simple task to change a domain name for an instance without breaking federation, so once you have a federating instance you are are stuck with that domain name. I would not recommend picking an .xyz domain for your instances unless you intend it to be a private instance forever. If possible, try to go for a .com, or some other more standard TLD!

Is it in bad taste to take content directly from Reddit/Stack exchange?
I would like to make a community dedicated to posting interesting science questions that have been asked and answered in other sites. The post would be a copy of the question and the comments would be copies of the answers that were found to be interesting - always with the username attribution and the link to the original post. With this format, users would not need to leave the site, and it is easy to discuss the answers directly in the comments. But I have my doubts about whether this is appropriate, as I think that it might be copyright infringement. So, what do people in Lemmy think? Would this format be blatant theft and wrong? Perfectly reasonable? Somewhere in between?

Using a VPN provider vs self-hosting a VPN
I have seen the following argument (summarized here as I understand it): *Despite the promises that VPN providers make, it is known that they will often monitor your traffic, collect logs, might share your information, and will collaborate with law enforcement. Renting a VPS and running an OpenVPN server on it and using that as your VPN, is better - because you have full control over the logs. Let's assume we trust the VPS provider to adhere to their TOS and privacy policy.* To talk about a concrete typical usecase, I am thinking about how this applies to downloading illegal torrents. In my current view, the only scenario in which the self-hosted option makes sense is if you pay for hosting using crypto and reveal no personal information during the process. Otherwise using a VPS would be virtually the same as downloading it through your ISP - and in some cases even worse - because the VPS provider might be more easily pushed to throwing you under the bus if abuse is reported since this might be a TOS violation. On the other hand, a VPN provider has a much larger motivation to protect users against this because the way that users perceive these protections is fundamental to their business model. So, is there a reason to self-host a VPN instead of using a VPN provider? If so, should the VPS be acquired anonymously, or are there ways to protect yourself while using a provider that you gave your personal information to?

Have you experienced ASMR?
'[ASMR](' - or 'Autonomous Sensory Meridian Response' exploded in popularity a few years ago as the ASMR youtube channels emerged. The YouTube videos do not trigger ASMR for me, unfortunately. However, it did bring the term to my attention, and when I read the wikipedia page something clicked - I finally had a term for a very strange sensation that I was never able to explain to others. For me, this ASMR response is not something that I can reproduce on purpose. But the times that it happens it usually occurs when someone else asks me something. For example, during middleschool I remember that sometimes a friend would stand next to me and ask me for a pencil. As I would retrieve the pencil, I would get into this dissociated-like state that was very peaceful and tingly, and I would not want it to stop. I would be very slow in retrieving the pencil and pretending I was still looking for it - and then I would ask some question to try to make the interaction longer because it felt so interesting. It still happens some times but is rare, and it occurs when I am busy in the office or the lab and someone comes and asks me something or for something while I am paying attention to something else. It is a very pleasurable state that is also very fragile - as if someone managed to scratch the perfect itchy spot in my brain for a moment. It is odd, and I think it fits the definition of ASMR perfectly. The only other way that I feel a similar effect but not as strong effect is when am getting a haircut the hairdraisser is cutting the hair in the back of my neck with the tiny trimmer. I am curious about other people's experience with this phenomenon, because I have not found many people that can relate to my experience. Have you felt this? Do you know what your triggers are? Do the YouTube videos do it for you?

How have you optimized your Pi-hole?
A few months ago I set up a Pi-hole at home. I tried to create a decent blacklist by first adding some regexes I found online and then doing some manual optimization by finding ads and blocking the ad requester through the queries page. But in the end it did not work so well. I still get a lot of YouTube ads, for example. So I still rely on uBlock for my ad-blocking. I just read that Chrome is planning to start weakening ad blocker plugins. I use Firefox, but it is still worrying that the space could potentially keep moving in this direction over time. So I am now more invested in learning how to have more control over my ad-blocking. Has any of you managed to optimize your Pi-hole to the point that it is possible to browse the internet without an ad-blocker? If so, how did you do it? Are there other good ad-blocking alternatives?

[PDF] Aging-associated cognitive decline is reversed by D-serine supplementation
Summary D-serine, which acts on a type of N-methyl-D-aspartate receptors (NMDARs) is thought to be important for neuronal plasticity. As people age they produce less D-serine, and it is hypothesized that this is one of the reasons why it becomes more difficult to learn as we age. D-serine can be absorbed in the digestive tract and it can cross the blood-brain barrier. It is hypothesized that the D-serine levels of the aged brain can be restored via oral supplementation, reversing the cognitive decline to a degree. The experiment used to test age-related learning deficit is the following: a rat is trained to press one of two levers to receive food. Then, the experiments switch the correct lever, such that now the rat needs to press the other lever to get the reward. The cognitive flexibility is tested by measuring how long it changes for the rat to begin pressing the other lever. Young rats learn to press the other lever quickly, while older rats continue pressing the old lever for longer. When supplemented with D-serine, the ability of the old rats to re-learn which lever to press improves dramatically. They show that D-serine supplementation has no effect on the young rats' performance, and they show that this effect is dose dependent, supporting their original hypothesis that D-serine supplementation can restore some of the brain's plasticity that is lost as the rat ages.

Have you had success correcting your bad posture?
I slouch a lot, so my shoulders are a bit rounded at my neutral posture. I have tried to get into the habit of doing wall glides, stretching my chest, doing back exercises, and sitting properly. But after quite a few years of testing strategies, tools, and trying to build consistent habits, I usually manage to get two or three weeks of good progress but then fall back to the usual bad habits and my posture does not really improve. Do any of you have any success stories about correcting your bad posture? I would like to learn from your success!

Is there a simple way to remove a banned user’s profile description?
A user was created in my instance to spam phishing links to some google sites that look like bank login sites and crypto wallets. I banned the user and removed their content, but the links were also added to the description in their profile page, and they still live there. Is there a way to delete the description through the UI? The only way that I can think of how to do that is to manually remove the description in the postgresql database.

E-mail server configuration
Hello! I have procrastinated configuring the e-mail server for my instance. I am looking into it and it is not entirely clear to me how I should go about it. I see that some docker-compose files have the following block added: ``` postfix: image: mwader/postfix-relay environment: - POSTFIX_myhostname=${domain} restart: "always" ``` But I don't understand whether this is supposed to work out-of-the box (which it did not for me when I tried), or if I need to set up the postfix server as one usually would and the postfix-relay somehow makes the connection between docker container and the already configured server. Is there a guide for complete beginners? I could not find one in the lemmy docs and I have never configured an e-mail server before. I am reading through the postfix documentation and already got some basics but there is a lot, and I am not sure that I am moving efficiently towards the end goal 😅

Would you be interested in being able to recover your forgotten password using PGP?
I was thinking about how one could go about recovering their forgotten password without providing an e-mail address. One way that I can think of is to have a section in one's profile for adding your GPG fingerprint. If you click "forgot password", the server can generate a code, encrypt it against your public key, and provide you with the PGP encrypted message block. You would then use your private key to decrypt the message, input the code, and change your password. I have not seen this type of password recovery being implemented in other sites, and I am not sure why. Is it that people are just not interested in having this option? Or maybe those who can keep a their private key secure are also usually able to keep their passwords secure and well-managed?

Personal attacks decrease user activity in social networking platforms (More specifically, Reddit)
[Link to full PDF](

[Question] VMWare vs vStack cloud server
I am currently running my instance using a VMWare cloud server with the following specs: 1 GB RAM / 1 Core / 30 GB SSD / 10 Mbps This costs 13.94 €/month. Using the same provider (, I can rent a vStack cloud server with: 1 GB RAM / 1 Core / 30 GB SSD / 50 Mbps For 4.55 €/month. I am using the more expensive option because this is my first self-hosted project and I did not realize the cheaper option was available when I originally set up the server. As I currently understand it, renting a VMWare is renting a physical machine, and renting a vStack is renting one of many virtual machines installed within a physical machine. Is this correct? And, are there any practical reasons not to "downgrade" to a vStack? When should someone choose a VMWare over a vStack?

Firefox appears to be down
I am not able to load any websites with Firefox on my PC despite having an internet connection. I went on Reddit and found other people complaining of the exact same issue, so it appears to be a problem with firefox: My phone's firefox browser works fine. What can cause a browser to fail this way?

Are there any flaws or problems with Nano?
I have been reading about cryptocurrencies, and I find Nano to be the most appealing on the basis that it appears to be a minimalist approach to digital cash - which is what I would like. I want to be able to pay online easily without needing a third party. I don't really need smart contracts or any other fancy crypto stuff to be tied directly to the digital currency. The mechanism of decentralization via Open Representative Voting appears to be much less wasteful than the competitive PoW algorithm that bitcoin uses, so that is quite sensible. I was concerned about the concept of the public ledger allowing others to look into your wallet and trace back transactions, but learning about trust-less mixing has dissipated those concerns. I see people speaking wonders about Nano in nano-related forums (of course), but outside of those I get the feeling that people either pay little attention to it or generally dislike the currency... I have read many claims about spam attacks, but those appear to have been dealt with successfully looking at the recent updates and discussions on the official forum. So, are there any obvious negatives that I am missing?

How can I serve a static file in response to a Webfinger query?
I am trying to follow [this ActivityPub tutorial]( I would like to get my server to respond with a static file when it gets a query for Since it is not explained in the tutorial I suppose that this is a very basic thing to do, but I have never dealt with this type of query before. What is the simplest way achieve this? My server is using nginx to serve a basic static html at

Etiquette for git contributions
I am learning how to use git, rust, and the ActivityPub protocol. I am looking through issues in github and trying to see if there are issues that are simple enough to help with. I have read the ["Contributing"]( section, but it is not very specific. I have a few questions: - Should one discuss with developers before submitting a pull request, or is it better to submit a pull request and then they can decide if they accept it or not? - Should pull requests be sent to merge changes into the master branch, or to a different branch? - Is it ok for users to make small contribution, such as fixing a typo or small bug? Or should users only attempt to contribute when they have a more significant contribution to make, such as implementing federation with a new project? I have found many resources about how to work with Git but not about the etiquette. So if anyone has some resources on the etiquette for collaborating to projects via github, or specifically for Lemmy, that would be very helpful!

I have been learning a lot thanks to Lemmy and the people behind it, so I wanted to contribute a bit back by making some easy-to-follow video tutorials. I have also made the same video in Spanish: As a beginner myself, I do have a level of concern that I might teach something incorrectly, or that I might expose others to security risks and liabilities. I hope that these concerns are not well-founded, but I am open to criticism!