Slack updates retention policy for free workspaces
slack.com
slack.com
For private messages I think you have to send in a GDPR request.
You can view it in a slack like viewer here: https://github.com/hfaran/slack-export-viewer
Also note that you already can’t see messages older than 90 days on a free workspace, but they’re there if you upgrade. This change means if you do upgrade, data older than 1 year will be gone.
I think the approach will be to upgrade, export the data and then cancel subscription.
I don't mean cancel as to leave slack but to be able to revert to the new free usage policy while keeping previous data.
That said, there can also be a large body of fruitful technical knowledge sharing that happens on these platforms that should be retained for the good of the company.
The real rub is throwing the baby out with the bathwater. How can the signal be isolated/retained/copied easily so the noise can be purged? This is especially difficult for very old data that is spread across innumerable channels whose value could be lost and the team left unawares.
I've never liked Slack/Discord for anything other than instant messaging, and always thought of it as ephemeral. If it's anything of significant importance, it should be documented in a place that is meant, by design, for documentation.
The challenge we have found is that amazing technical documentation is often born from ephemeral conversations. I would love a solution that eliminated the toil required to clip out those comments into a long-term store.
The flip-side to the challenge is that moving our team from Slack to another platform better suited to long-term technical Q&A requires a significant cultural shift that may not be practical.
I think you need a culture of summarising this type of info & saving somewhere that isn't ephemeral (e.g. README, team wiki). Else even with the history maintained, it's still a pain to try to a) play pattern matching with Slack's search engine and b) weed out all the false positives of old info that's no longer relevant.
This feels like the type of thing that maybe could be farmed out to a trained LLM in the future — "summarise today's discussion on X in channel Y and post into the wiki".
Record retention is a reality for many workers no matter how inconvenient reading out of context messages from 7 years ago might be.
Since then the game has been: use investor money to pay for servers and collect users, then sell them to some big company.
An actual profitable chat service, like, as a business, can’t exist I think. It is impossible to compete with free service provided for “free” (free as in barnyard animals don’t have to pay for their dinner).
Lest this come across as smug or whatever, I’m personally eating from the Discord trough. It is what it is.
IMO, it is, as long as you don't need video sharing or mobile. As soon as these are involved, the field of competitors narrows considerably.
It's not that hard to click 'save'. The old needs to make room for the new.
Perhaps a better way to state my position is: I think publishing should be a ratchet. Everyone can choose what they want to make public. But once it's public that's it, there's no undo and no do-over. This isn't ideal in some situations, but on balance seems like the greater good. Stuff gets added to the commons of human knowledge, and we do our darnedest not to let stuff get removed once it's been added.
20 or so years back, I joined The WELL. I only lasted a couple months. It was impossible to have an interesting conversation without longtime members popping into the thread to "helpfully" point out that this had already been discussed back in 1988, 1993 and twice in 1997, with links to the relevant threads. I figured, if The WELL wants to be an information archive rather than a social space, that's fine, but Wikipedia is less expensive, better-curated, and easier to search.
About 15 years back, I gave up on contributing to Stack Exchange, and about 5 years back I largely stopped clicking on SE links in my search results. Similar story. The culture of preservationism meant that the site tended to favor pre-existing information and opinions. Back in the day, it just made the site uncomfortable for newcomers. More recently, it's also undermined the site's merit as an knowledge trove due to a tendency to favor out of date information.
Also about 15 years back, I became an admin on a Web forum. I observed firsthand that having a searchable record of posting history was a social negative that largely served to fuel flamewars. More Internet-savvy users were very adept at searching through the post history to dig up dirt on each other, hunt for evidence of people's opinions changing over time in an effort to try to call each other hypocrites, etc.
The rise of social media has only made this worse; Twitter mobs' ability to dig through a target's history for ill-considered or easily-misconstrued tweets that they can use for character assassination purposes is truly a sight to behold. And it has created a situation where more level-headed, continent people tend toward circumspection or even avoiding posting altogether.
On the professional side of things, at work I find that the usual corporate information troves - Confluence, ticketing systems, and group messaging - are rarely useful as an information trove, precisely because we hoard everything. It makes actually finding the information you're looking for in these systems darn near infeasible. Sturgeon's Law ("90% of everything is crap") is only true in the short term. In the longer term, 90% of what remains is compost, because it only ever had ephemeral value. Digging the reasoning behind a 5 year old technical decision out of a 10 year old Jira system is like looking for a chocolate chip in a cesspit.
Unfortunately, we've collectively painted ourselves into a corner in this department. Once upon a time, we curated information because we simply didn't have the physical space to keep all that paper indefinitely. Cheap digital storage and keyword search has tricked us into thinking that there's no cost to storing everything, though, so over the past quarter century or so we've gotten rid of all the secretaries, librarians and archivists.
On the surface this just looks like Slack saying “we used to keep everything in case you upgraded but now we just keep a year back”.
Functionally it’s no different if you never planned on paying. If you did plan on paying then now you won’t get all your messages from the past back, just a year.
Have I misunderstood this?
> To improve our services. We use your information to help us understand how users interact with our services, what features or products users may want, to develop features that make Discord safer and better to use, or to otherwise understand and improve our services.
> you can limit certain types of processing of your information:
> Limit our ability to use your data to improve our services. We offer certain settings that allow you to limit the information we collect to help us understand how users use our services, what features or products they may want, or to otherwise help us improve our business. If you turn off these settings, we will stop collecting and using certain event and log information as described in these settings.
It also majorly creeps me out that everything I've said in what's supposed to be a casual conversation is searchable forever. Not that deleting old messages would necessarily change anything, but I don't see keeping them as a good feature.
I mean, I thought after 2 years you're supposed to delete the data, assuming the account is inactive. Google started doing that and I imagine if your old long-forgotten account got hacked and there was messages/attachments from 2+ years ago they would be fully liable as data is not supposed to be held for longer than reasonable. 10 years can be considered unreasonable in most circumstances.
But I cannot fathom why they want to pay for this, it's basically an unlimited storage file hosting platform they must be hosting TB's of probably useless data.
What data are you expecting Discord to delete?
Not everything is PII.
‘Personal data’ means any information relating to an identified or identifiable natural person (‘data subject’); an identifiable natural person is one who can be identified, directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data, an online identifier or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural or social identity of that natural person.
https://support.discord.com/hc/en-us/articles/360003858092-P...
As for messages, how can one know if each and every message may or may not contain PII? Or an attachment? You can't, which is why the best approach is a cautious one:
Delete all data from inactive accounts after 2 years, or even earlier to limit liability. The GDPR dictates you only store which is strictly required, messages from 10+ years ago from accounts that have not logged in for years have no right or nesscessity in being there
https://x.com/stewart/status/780906639301812225
Via Wikipedia, etc.
It just isn't that valuable, even without the huge amount of negative publicity attached to doing that.
The cutting edge AI labs are leaning much more into high quality data (licensed from the Associated Press for example) and synthetic data, which it turns out is a huge part of Claude and Microsoft's Phi series.
Andrej Karpathy said: "The average webpage on the internet is so random and terrible it's not even clear how prior LLMs learn anything at all." - https://twitter.com/karpathy/status/1797313173449764933
Can you imagine the storm of bad publicity that would emerge the first time some company has details of an internal strategy leaked because some chatbot ended up parroting those details back to a competitor?
I get deleting them for little used workspaces, but if they are free and used, this should not happen.
I personally put the downloaded zip file through https://github.com/akkartik/slack-archive. But there are a variety of tools that can read the zip file.