Self-plug: consider Handmade Cities. We have a simple meetups [0] page if you decide you appreciate our ethos. Hopefully we have an active meetup location near you?
In any case, good luck on finding the right community!
One of the difficulties of public relations is communicating to multiple audiences at once. One of Docker’s audiences are the paying customers who outside that 2% and would want some assurance that if docker makes errors, those errors are smaller in magnitude. This statement seems like it is aimed at assuaging the worries of that audience. Is it good practice? I do not know.
Is that a serious suggestion? Who on earth would take the time to look through all of those to see which one best fits them? Companies have enough trouble getting press releases read already.
I guess then you could use some heuristics and ChatGPT to tailor the press release to the particular viewer.
For example if the user has a screen resolution of three 4K monitors side by side, using Linux, and coming from a Silicon Valley IP address, they are probably a developer. If they have the screen resolution of an iPad Pro and a New York IP address they are probably an executive on the go. The HTML5 accelerometer API might also say something about whether they're reading your press release in bed, while sitting, or standing. Use ChatGPT to reword the press release appropriately.
This is a genuinely terrifying vision of the future. Already things like prices may differ based one’s location. Adding content writing AI to the mix is positively dystopian.
What if you are a "closed source developer" at work but an open-source one during your free time? What if you are a billionaire codes? What if… etc. You could make TL;DR's targeted at specific audiences, but you still need the same introduction for everyone.
The 2% they’re referring to are businesses that are using Docker’s hosted services for free. The majority of the outrage was from people thinking about the non-business users, that is, open source projects, which Docker unintentionally implied would be impacted by this change. Docker are apologising for their poor communication which made people think this change applied to more than just a tiny portion of the user base (who are probably happy to pay). They’re not apologising for the change.
Anybody who uses "docker pull" or "FROM" and not pointing at their own hosting or their own paid Docker account was affected as evidenced by the thousands of comments worried about the impact.
> We’d also like to clarify that public images will only be removed from Docker Hub if their maintainer decides to delete them.
> Will open source images I rely on get deleted?
> Not by Docker. Public images will only disappear if the maintainer of the image decides to proactively delete it from Docker Hub. If the maintainer takes no action, we will continue to distribute their public images.
People may have thought they were affected, which is what they seem to be apoligising for.
They also are saying the maintainers will be unable to update the images after the 30 days. So the panic and bitching are perfectly deserved: https://news.ycombinator.com/item?id=35188691
For TEAM accounts that aren't "Docker sponsored open source" teams.
They should allow a TEAM->PERSONAL conversion for any open source account that doesn't qualify to be "Docker sponsored." But really this is a communications fail more than anything.
This only ever applied to the *Team* accounts. I have a paid non-team/personal account, but I am also aware that I could have a free personal account if I didn't need private repositories.
In other words, they weren't clear enough in their communication, which is what they're apologizing for.
But the internet outrage mob is going to yell about the evil of The Man no matter what I say, so I don't know why I bother...
> Less than 2% of Docker users have a Free Team organization on their account.
I don't think so. The quote above is what they say on that page, and I think that is a pretty useless metric. It affects 2% of all Docker Hub users, 100% of all Free Team users.
Ever wonder why Google outage notifications always say stuff like "this impacted 0.01752% of users"? Because if they leave that out, the PR department ends up flooded with questions from reporters about "how bad was this outage, exactly?", and less-diligent publications running "Google suffers massive outage" headlines.
It's really misleading though, as it only reflect the owners of the images. Presumably I should count as an affected user if I don't own the image, but try to download it.
> If you don't upgrade to a paid subscription, Docker will retain your organization data for 30 days, after which it will be subject to deletion. During that period you will maintain access to any of your public images.
That sounds a lot like the public images were subject to deletion. At the very least, subject to being frozen in time and not updated/updateable, which can be worse in some cases.
It's worse than just handwaving, it's straight up nonsensical. Perhaps it is literally true that only 2% of distinct accounts that log into Docker Hub have this plan, but for the vast majority of people "using Docker Hub" means "pulling public images from Docker Hub", not "logging into Docker Hub", so by a more reasonable criteria (say % of images pulled) I'm sure its at least an order of magnitude greater.
Yea. I don't think its necessary to even add that part since it seems the only reason they are responding is due to the negative backlash from the open source community. The fact that they added the % appears to me as an attempt to marginalize and compartmentalize the perception of this decision so it doesn't spread into the perception of their paying customers. Good luck with that.
(Note: I'm neutral on the main issue of whether Docker's moves are evil, etc. I don't really care)
To me this "This impacted less than x%" business is more of a classic Apple damage control PR statement, designed to convey to the whole userbase, "You almost definitely aren't affected, it's just a tiny number of whiners making all this fuss, and look how small they are!"
I think that was an attempt to describe that the change won't delete tons of projects like many believed, and break downstream users, not that it was too small of a group to care about.
Also, the 2% of the impacted users might have 50% of _all_ users as a dependency (just throwing out a random number for illustration), so I'm not sure that the "2% users" messaging matters to the recipients of that PR.
> 'We're sorry we mistreated you, look how small you are to us.'
They could choose not to share any data, which is what most companies default to.
You're complaining about something so small as if they aren't handling this entire thing beautifully at this point. They noticed their mistake, and corrected it swiftly to keep the community from bifurcating. What else do you want, exactly?
No there isn't. This is entirely subjective and you're acting like they said "Fuck our customers" when they just shared data. Anything you want to imply beyond that says more about you than it does about any part of Docker.
> The first implies that they have up to 2% of users which they don't respect, and undermines their apology.
Where does this implication come from? Why is Docker not given the benefit of the doubt when they are already extending an olive branch...? This isn't Microsoft.
I guess if you want to change things, you should shoot for a position in PR at docker. Otherwise, you look like a rube for acting as though they "could have done better with one sentence." I bet you're fun at parties.
They said they intend to delete customer data if they don't switch to paid plans: this means "fuck our customers", collectively and indiscriminately, not only the selected foreclosed accounts but everybody that depends on them.
> They said they intend to delete customer data if they don't switch to paid plans
Customers who are running businesses and knowingly breaking the ToS? I'm not sure why businesses like Docker aren't allowed to defend their revenue.
It's actually pretty hilarious how many of you are coming out of the woodwork to attack Docker, they are not the enemy in any way, shape or form and if they disappear you're gonna miss them.
Personally, I expect what I paid for from a free service, and I consider depending on gifts and uncertain platforms a reckless risk.
But, as a business whose prosperity depends on the goodwill of masses of users, Docker can and should "defend their revenue" in a way that minimizes collateral damage in the form of
1) gratuitous bullshit, untrustworthiness, lack of transparency, and perceived evil intentions (e.g. their second announcement)
2) technical uncertainty and security risks (for orphaned images of uncertain status)
3) inconvenience, without additional revenue, to the vast majority of users that aren't included in this the shakedown
There are rational businesses and there are businesses that drive them away their customers; in the long term, the former tend to "defend their revenue" much better.
It feels like people just enjoy digging into outrage - this sort of nit picking at specific phrasing in communications is baseless, and happens for literally any incident where an apology is issued.
Its wild how the same people will complain that some corporate missive is completely content-free while at the same time punishing any attempt at earnest communication by scouring the missive for a raised edge to take offense at.
The 'problematic' phrase is right before the section detailing who is not impacted - which is what many people are going to want to know! "Wait, should I be worried about this?'
This is an example of my point actually - these types of posts are magnets for people that cannot be pleased.
Agreed. It's frustrating to see companies try to downplay the impact of their mistakes by using statistics like "less than 2% of our users." Companies NEED to take responsibility for their actions and show genuine empathy for those affected, rather than trying to minimize the impact of their mistakes with vague statistics.
The "2% of our users" is indeed misleading since most users don't run organizations, and it's mainly orgs that were affected. A better metric would be what % of orgs were impacted.
