@PhlipElder said in Staying in Ethics and Legal with ChatGPT usage?:

Being able to stand back, look at something, and say, "I did that!" is a pretty amazing experience. So, did none of the folks bastardizing their writing ever have that experience?

Yes, but college isn't work. It's busy work. The entire point of university isn't education or work, but to buy a degree. ChatGPT is the logical path to that. If the students (or their future employers) valued education, they'd test for that. As long as they prefer a piece of paper over actual learning, there is no dignity in the process.

@PhlipElder said in Staying in Ethics and Legal with ChatGPT usage?:

Have we really gone that far that "grey" justifies virtually any kind of behaviour?

Try it in reverse. Try to justify to me being a PhD student or an employee and having access to ChatGPT and not using it. As an educator and employer, I see avoiding the use of the available tools as lazy and wrong. If you feel that it is justified to excuse the grey area of doing manual work where none is needed and doesn't add value (in my estimation) then explain to me the opposite.... how do you even excuse not using the available tools and just filling the student and/or employee's time with pointless busywork?

@Obsolesce said in Staying in Ethics and Legal with ChatGPT usage?:

That sounds more like the kind of situation being those people would have gotten infected just the same regardless of web browser used. Latest version of the web browser prior to infection?

As far as we can tell. It's on managed systems that are automatically updated, AV is up to date and active, firewall is on. But just takes clicking on something.

We project that Edge puts people at additional risk because it is the default product on the most insecure platform, that is also a default choice. It makes it super likely that your target is "accepting everything because it is default" rather than being thoughtful in their technology choices. It makes it an ideal public "flag" to make someone a higher than average potential for malware.

@PhlipElder said in Staying in Ethics and Legal with ChatGPT usage?:

I've come to realize that will never again trust anything from anyone without proof that they know what they have presented.
That's going to be the differentiator in my mind.

Ah, but I think this just exposes something big. Why did you feel that they knew the material before? This is why PhD students defend their thesis... anyone can produce the paper, it's explaining and defending the concepts live that get you a degree, not the paper.

If you were using the written paper as a proxy for testing someone's knowledge on something then yes, I can see why you care about the mechanism rather than the output. But I'd say, again, all that is happening is that what was already true is being exposed.

As someone who made a living for a while in high school writing essays by request, I know how common it is to not have written your own paper. I don't know why people bought essays from me, whether they used them as source material, cited them, used them to summarize research, or turned them in as their own, not my concern. I was hired to write papers on topics. I didn't even know who got them. But I know actual intelligence went into writing papers that were used by people who knew nothing of the material.

When I went to university, the top ranked uni in the US at the time that I went, it was expected, that you had copied answers from previous years. They assumed what other places called cheating as a baseline and tested only above that. If you didn't take the time to obtain and memorize previous years tests you would almost certainly fail. They didn't test only on that, they assumed it as a baseline of available knowledge.

So I see what you are saying, but what I'm saying is that the inability to trust that producing a paper that has good words on it to reflect on the knowledge of the person turning it in was already there. ChatGPT isn't changing the game there, in any way. Authors of works, even if they wrote every word themselves, rarely understand the material deeply. Writing an essay simply is not a good test of that.

So the issue, and the solution, should be pretty clear. Essay writing was not ever a great process for education (or work), we've just exposed it beyond question now. But for many of us, that happened long, long ago. Now you need to focus on discourse, which has always been the case.

@stacksofplates said in Beelink PC issues:

I've bought a couple of the micro form factor Optiplex computers (9020) and have been happy with them. You couldn't have saved too much by buying something like this I can't imagine? I think I paid $250 for the last one and it came with 8GB RAM, an i7, and a 250GB SSD.

The 9020 lists as running a Four Generation processor. That's eight generations old. That's dramatically old. Ten years, in fact. Those came out in early 2013.

Similar price for high end, brand new genuine AMD fro Beelink. The fact that it died sucks, so that's a real issue. But the two aren't comparable in performance.

Maybe you have the wrong Optiplex number of Dell has the wrong info on their spec sheet? That looks like a $20 computer to me....

https://i.dell.com/sites/doccontent/shared-content/data-sheets/en/Documents/optiplex-9020-micro-technical-spec-sheet.pdf

@PhlipElder said in Staying in Ethics and Legal with ChatGPT usage?:

@scottalanmiller said in Staying in Ethics and Legal with ChatGPT usage?:

Every generation spends time learning new tools. Every old generation feels like this is lazy. But just as we use printers instead of type writers. And our grandparents used typewriters instead of pen and paper. And their great great great grandparents used pens instead of chisels and rocks... it's not that we are increasingly lazy. It is that we are able to reduce the amount of wasted effort so that we can spend more time on the parts that are important.

Today we can write more intelligent discourse and communicate about it in minutes than stone age man could record in a month and share with no one. The use of tools to eliminate or reduce the unnecessary allows time to focus on real learning, growth, and productivity.

Just remembered another example: Cole's Notes.

Papers written based on a reading of Cole's Notes as opposed to actually reading the book, absorbing it, understanding it, and then being able to see the author's intent would be very different.

I want real the real person and their real experience.

https://open.spotify.com/track/73CKjW3vsUXRpy3NnX4H7F?si=ac956108dbb54fd4
^^^
Fake Plastic Trees

If reading Cole's Notes produces enough to write a good paper it tells us one of two things...

1. The notes are as good as the "real thing" or...
2. The evaluation of the assessment is bad and pointless.

If the goal is to enjoy the material, you don't need to be tested on it. If the goal is to pass a test, the notes are normally vastly superior.

If you know the goal, you can decide which is the better approach. If doing this for school and not for personal growth (the two are opposing concepts) then Cole's Notes are much better as the time spent reading something not good enough to read without it being assigned is wasted and the value is in passing the assessment.

@dashrender said in Remembering the MCSE+I, Microsoft's Terminal Certification:

the NT 4.0 workstation books left out all kinds of unix/linux connection/printing part meaning you didn't stand a chance on those parts on the workstation test unless you had external knowledge.
I barely passed the workstation test because of this.

I never took the NT4 Workstation. I did the WIndows 98 / DOS test instead, the one where the study book was over 2,000 pages. It was harder than the rest of the MCSE+I combined. It was definitely the foolish choice as it was easily 20x the work of the NT 4 Workstation test and not one single item of all of that did I ever use, ever, in my entire career.

@Cagatay said in Remembering the MCSE+I, Microsoft's Terminal Certification:

Lol here is mine... does it still have value these days? or are we considered as dinasours?

Oh man, I don't remember getting a cool card like that. But it does look a little familiar.

That's very cool that yu still have it!

@WLS-ITGuy said in Debian 11 & php8:

One of the applications we use just released a new version and the update requires php8.0 or above.

We're using Debian 11 and since 11.7 was just released, which doesn't have php8 in the release. I was wondering how do I find out when things like php. Mariadb, Apache, NGNIX, etc get applied to distros?

If you want any kind of modernity, Debian isn't really for you OR you use Debian as a base and do not use it as your package testing and repo system - which is generally not advised in production, but it is an okay approach as long as you accept it. Basically it means you are using Debian as a base and assembling your own distro instead of trusting the vendors.

With my CIO hat on, we never do that. If we want modern software, and we normally do, we run Ubuntu or Fedora. Both of which have had PHP 8 and 8.1 (8.2 is current) for quite a long time. Debian is great as a base and when you want things that never change. But it is not good when you want things that are keeping up to date.

@Pete-S said in Debian 11 & php8:

Not a challenge at all but the reason to run "stable" is for stability.

Once you start abandoning the integration, though, you are abandoning stability. The idea of using an LTS and then replacing the parts of the OS that aren't up to date is counterproductive. Choose the most up to date, best supported, most stable version and use the fully tested and integrated components instead.

The idea of "stable" is not stability in IT terms, that's a myth. It's actually against that. The idea of current is for IT stability. Stable, in reference to an OS like this, is in reference to the versions of products remaining stable so that unsupported, out of date software from bad vendors can be used without updating for long periods of time. Not a positive stable, it's a bad stable.

I think what you want is a completely external device. That's how this is normally handled. Meaning it's common to have a dialer / ringer on a computer but you answer the phone. Same thing could be done to make a loudhorn blast anything you want as well.

Mostly this is taken from Mastodon's documentation but some of it is wrong. I tried their Docker configs, but couldn't find any that worked and it isn't clear if they have an official Docker image or just third party ones. They don't actually list Docker on their official site. So this is a traditional install.

As root...

apt update && apt install -y curl wget gnupg apt-transport-https lsb-release ca-certificates
curl -sL https://deb.nodesource.com/setup_16.x | bash -
curl -sL https://dl.yarnpkg.com/debian/pubkey.gpg | gpg --dearmor | tee /usr/share/keyrings/yarnkey.gpg >/dev/null
echo "deb [signed-by=/usr/share/keyrings/yarnkey.gpg] https://dl.yarnpkg.com/debian stable main" | tee /etc/apt/sources.list.d/yarn.list
wget -O /usr/share/keyrings/postgresql.asc https://www.postgresql.org/media/keys/ACCC4CF8.asc
echo "deb [signed-by=/usr/share/keyrings/postgresql.asc] http://apt.postgresql.org/pub/repos/apt $(lsb_release -cs)-pgdg main" > /etc/apt/sources.list.d/postgresql.list
apt update && apt install -y imagemagick ffmpeg libpq-dev libxml2-dev libxslt1-dev file git-core \
  g++ libprotobuf-dev protobuf-compiler pkg-config nodejs gcc autoconf \
  bison build-essential libssl-dev libyaml-dev libreadline6-dev \
  zlib1g-dev libncurses5-dev libffi-dev libgdbm-dev \
  nginx redis-server redis-tools postgresql postgresql-contrib \
  certbot python3-certbot-nginx libidn11-dev libicu-dev libjemalloc-dev nodejs
corepack enable && yarn set version classic
adduser --disabled-login --gecos "" mastodon
sudo -u postgres psql -c 'CREATE USER mastodon CREATEDB;'
ufw allow http && ufw allow https

As Mastodon

su - mastodon
git clone https://github.com/rbenv/rbenv.git ~/.rbenv && cd ~/.rbenv && src/configure && make -C src
echo 'export PATH="$HOME/.rbenv/bin:$PATH"' >> ~/.bashrc && echo 'eval "$(rbenv init -)"' >> ~/.bashrc
exec bash
git clone https://github.com/rbenv/ruby-build.git ~/.rbenv/plugins/ruby-build
RUBY_CONFIGURE_OPTS=--with-jemalloc rbenv install 3.0.6
rbenv global 3.0.6
gem install bundler --no-document
cd ~
git clone https://github.com/mastodon/mastodon.git live && cd live
git checkout $(git tag -l | grep -v 'rc[0-9]*$' | sort -V | tail -n 1)
bundle config deployment 'true'
bundle config without 'development test'
bundle install -j$(getconf _NPROCESSORS_ONLN)
yarn install --pure-lockfile
RAILS_ENV=production bundle exec rake mastodon:setup
npx update-browserslist-db@latest
exit

As root again...

cp /home/mastodon/live/dist/nginx.conf /etc/nginx/sites-available/mastodon
certbot --nginx -d  yourdomain.com
ln -s /etc/nginx/sites-available/mastodon /etc/nginx/sites-enabled/mastodon
vi /etc/nginx/sites-available/mastodon
rm /etc/nginx/sites-enabled/default
systemctl restart nginx
cp /home/mastodon/live/dist/mastodon-*.service /etc/systemd/system/
systemctl daemon-reload
systemctl enable --now mastodon-web mastodon-sidekiq mastodon-streaming

This is tested with Mastodon 4.1.2. ZeptoMail is Zoho transactional email system. This is not well documented from either side, so here are the working details.

SMTP_SERVER=smtp.zeptomail.com
SMTP_PORT=587
SMTP_LOGIN=emailappsmtp.20ct9yn34098tv75 (this is your login, NOT the generic API that Zepto shows in samples)
SMTP_PASSWORD=password (the password supplied by ZeptoMail "less secure" option)
SMTP_AUTH_METHOD=plain
SMTP_OPENSSL_VERIFY_MODE=none
SMTP_ENABLE_STARTTLS=auto
[email protected]

@PhlipElder what are you using Mastodon for over there? Internal corporate use? Private group? Public server?

This is a testing instance for us, but we hope to take it live soon and it is going to be public.

@PhlipElder said in Configure Mastodon to Use Zoho ZeptoMail for SMTP Email:

We have an internal IIS or Exchange based SMTP server that we set up for this. It's just too painful otherwise to figure out.

We never do that because it's too temperamental. If you get blacklisted or anything, all it takes is your datacenter getting listed by some random group and suddenly O365 blocks you. We do this with Postfix which is faster, easier and more stable than Exchange (in places where we deal with Exchange we typically put Postfix in front of it for safety) but only to relay to ZeptoMail or SendGrid or Mailgun because that's how you make transactional Exchange or Postfix the most reliable.

Using FreshDesk here. Can't say that we love it, but it is definitely on the better side of things. We are on the free plan and it works fine. Lots of options and totally hosted for free.

If it is purely a single internal company, traditionally I've been most happy with Spiceworks. If you don't run up against their data configuration limits, it tends to be the best tech experience IMHO.

@Obsolesce said in Is it racist? I think it is.:

What's your evidence to support the site blocking is racially motivated, and not, lets say, regulatory or otherwise motivated?

Because no regulation anywhere, ever is supported by geo blocking. That never qualifies for any regulation.

@Obsolesce said in Is it racist? I think it is.:

Did you somehow miss the list I posted of so many more likely reasons a country is blocked?

None of those are even plausible. Let alone LIKELY.

@Obsolesce said in Is it racist? I think it is.:

geographic blocks are not racially motivated.

This is obviously backwards. As you've not, not has anyone in years of me talking about this, supplied any plausible or factual benefit to blocking, but the negatives, including loss of business, loss of reputation, etc. don't just make it anti-business, but generally illegal (fiduciary responsibility in larger companies.)