I disagree. I limit the screen time of my kids to 20 min per day. I only allow them also to view certain content that is not too stimulating, just educational stuff. The kids do want to talk about the stuff they see and play out what they see afterwards. I see it as positive. Unlimited youtube kids, with weird slop content, absolutely harmful.
yep, my 3 year old gets a very limited amount of screen time and he only watches educational programs (not whatever cartoons his peers watch). There's is no way I want to make it _easier_ for him to watch TV, especially as he has very little interest in it already.
Looking back in time, the only benefit of watching anything on a screen as a kid is learning a foreign language. The content is always some form of a brain rot, or political, cultural, or religious propaganda.
As bearish as I am on AI, outside of the agent deciding it only wants to play "Semi Charmed Life" and nothing else, I fail to see how this is a nightmarish hellscape in and of itself.
I feel standard commercial radio is already that hellscape. I mostly listen to rock (classic or alternative radio stations) and it's the same ~20 things being rinsed and repeated every hour and every day. Has been this way for decades.
I am pretty sure 90% of playlists you hear when out and about are generated by spotify algorithms at this point. I would more be concerned because it seems like a weird use for LLM when recommendation engines are probably better than it? I suppose it has a ton of training data around music forums and people recommending songs, but it is hard to believe that would beat out billions of hours of tracked listening spotify has access too.
That doesn't sound right. The hyperactivity can express itself in many ways, as you say – not necessarily physically.
Is your practitioner maybe really old? They changed the definition in 1987.
According to DSM‑5‑TR and ICD-11, ADHD is subdivided into three presentations: predominantly inattentive, predominantly hyperactive-impulsive, and combined.
Same here. For me the biggest bummer with GrapheneOS is that the promised new back up system is still not even on the horizon and was promised a gazillion years ago.
If you try to use the Slack APIs to scrape the data you will *quickly* run face-first into the insanely restrictive rate limiting they recently enacted to combat their customers using AI tools they aren't providing and able to monetize.
That being said, we were able to get full data exports in the past when we were merging two companies into a single slack instance. YMMV
When the org I was at moved away from Slack (due to costs) we used this method and wrote a little Python script to convert the main channels' JSON dumps into PDFs so we had a usable backup of channels.
Please do not include PDF and usable in one sentence. Setting up some simple Postgres with sonic for fuzzy search would be _usable_, but PDF is like migrating from Slack to Teams.
In this case we didn't need a long term solution for searchable data on Slack.
We did the migration in stages, basically this:
- Provide access to Teams
- Create all of the new teams / channels there
- Make Slack read-only but still keep the lights on
- Allow folks to search and reference historic data as needed with Slack
- Ensure everyone was moved over to Teams and felt ok enough using it
- Remove access to Slack
- Perform Slack export / PDF creation of important channels
- Attach Slack PDFs to important Teams channels
- Cancel Slack subscription
In the end, most people never even needed to use the PDFs because they got everything they needed out of Slack before access was removed, but they are there for peace of mind and a last resort.
We also took this as an opportunity to stop using chat as a source of truth for long lived information. Anything that should be stored long term made its way somewhere else (Jira, Confluence, etc.).
As a product, I like Slack a lot more than teams. Chatting is easier, the apps are better, reminders and scheduled messages, workflows, etc. Threaded conversations are much easier as well, and aren't artificially restricted to certain types of channels.
At a high level though, Salesforce is often seen as a predatory company. Prices are high and they will squeeze their users for every nickel (as demonstrated here). They will also monetize your data in ways that you probably don't want.
Teams chat is pretty bad, I agree. But it does have these benefits:
- Free if you're already in the Microsoft ecosystem, which most companies are.
- Microsoft is probably a better steward of your data.
- Teams video / audio calls are much much better than Slack huddles.
- That's pretty much it tbh
I'd bake them into a Sphinx static site. That gives you a free client side search index along with better navigability than sheets of paper. And you can target PDF if you still want it.
Well as per the article (and my own experience), the free tier only gives you public channels. The paid tier gives you everything: public/private channels, group chats (called MPIMs), and one-to-one DMs.
So yes, it breaks "privacy" (not that you should expect privacy when using a work Slack account).
IIRC, you have to do something called a "compliance export," which just like any other compliance feature (SSO, HIPAA BAA, audit logs, etc.) usually requires the highest plan. It's designed to add some extra friction so admins can't just add themselves to a DM from the main UI like they could with a channel, but it is possible.
We had dozens of channels with almost 10 years of business information in them.
Over time the business gravitated towards putting anything long lived into other sources but since migrating off Slack was essentially a kill switch on our data we wanted to make sure we had ways to access this historic data if needed.
There's no way non-developers were going to parse JSON files for text. We wanted a quick and dirty way to attach the archived PDF file for a channel as a file attachment to the new Teams channel. It gave everyone peace of mind that they could find anything later.
It all worked out in the end and was worth the few hours of dev time to make the 1 off script.
Btw I wasn't the one responsible for making the tech choice to use or leave Slack for Teams. I was the one who was tasked to help with the migration and help make things as streamlined as possible for the business to switch.
One of the biggest pain points was going back to a bunch of Google Drive, Jira, Confluence, etc. sources and finding + updating the links to Slack to be screenshots of the conversation. Another one was converting a bunch of Slack app / webhook integrations over. Teams is absolutely horrendous for this compared to Slack.
In what way? Maybe you have a different definition of human, but as one myself, and someone who produces lots of PDFs for others, it's absolutely readable. Double-click, open, read. Ultimately though the readability depends on the producer of the file...
I think you don't understand what "human-readable format" means. The Wikipedia article has a good overview[1]. Open any PDF in a text editor instead of a PDF viewer and you'll see why it's not a human-readable format.
Mattermost is great, we've used it at a few places and it's very flexible.
Extensibility and integrations with learning management systems, as well as owning all your data, makes it sound like a great option in particular for an education-oriented organization.
And I imaging the AWS or GCP costs for hosting it won't be as high as what Slack wants.
Used it for about 2 years at a growing startup. What I liked:
* Threads are way better than in Slack. They're top level instead of just a bolted on afterthought. This means all your various conversations scale way better and are way easier to find than in Slack. I can't overstate this enough, it's a killer feature and genuinely improves the overall organisation of your communications
* Font size is just slightly smaller. My eyesight isn't what it used to be, but I still think they get the balance of legibility and information density spot on, whereas Slack feels cartoony in comparison.
* Search felt a little better, I can't exactly put my finger on why or how, just that finding things in Slack always feels comparable to the iOS Mail search feature: very basic.
Drawbacks:
* There are less out of the box integrations I think.
You'd think so, but I've been hosting MM for about 6 years, and it's definitely gotten more user- and admin-hostile in that time. They've restricted really vital improvements to paying customers (basic stuff, like having a functioning search), removed existing features (like video calls) and started shoving in more advertisements and nags to upgrade to a higher tier.
The fact that they've ramped that stuff up so much in the last couple years does not bode well for the future, in my opinion.
My installation isn't associated with a business, it's just a chat board for about 30 people, so there's no question of me being willing to pay $300 a month for the privilege.
I'm sticking with Mattermost because there's no better option, and I've got hundreds of thousands of messages I don't want to lose. However, it isn't like they don't try to extort you just because they're better than Slack about it.
For what it's worth, Zulip has a Mattermost data import tool, and communities are an important use case when we set product direction.
While I can't promise we won't ever change our exact monetization strategy, we're not venture-funded, and thus are immune to the usual enshittification pressures. https://zulip.com/values/ has some context.
Is that using Jitsi or whatever it is? I thought that was third party? I actually set that up for work before realising that we all hate voice/video calls and would just prefer to type. :P
I hate having 300 apps on my phone (and I can't have them all on my home screen), but unfortunately, they're often much nicer to use than the corresponding websites.
They're much faster / snappier as they don't need to load everything.
Even if each click only takes a second to load or wait, it's annoying for many clicks.
Then there are all kinds of usability issues, like the page reloading in the middle of a form or process when scrolling up was misinterpreted as a refresh.
However, I'm not talking about apps that just load a webpage.
reply