The mods on Google+Help are consistently stating that Google can't export content from communities that was written...
The mods on Google+Help are consistently stating that Google can't export content from communities that was written by somebody else, even if you are the owner of the community. And that Google can't do this for reasons of privacy, GDPR, etc, etc. That you would need explicit permission of everyone in the community who has ever posted to save their content. And that this is why Takeout.G+communities only gives you a list of URLs for the posts.
This is beginning to really annoy me and so I I find myself compelled to argue back.
Here's what I wrote last time this came up in
https://plus.google.com/u/0/102282558639672545743/posts/icWeMWvj8hr
Technically, the G+ API let's you get access to any post and it's comments that you can view from code. That's the technique used in the open source tool linked above. You do have to jump through hoops to get a list of the posts in a community in a form that's useable by the API. So Google does provide tools to save the content. They just don't wrap it into an easy to use Community Takeout download for owners/moderators.
Procedurally, I've seen this work with no problems on other platforms. There's usually a public debate within the group/forum/community about where to go. That ends up with an announcement that the content is going to be moved asking for objections. And then if there are none, you just go and do it.
The request in the OP is a FAQ. There is very clearly a non-trivial need and desire to archive and migrate the content in some active communities. And that need and desire is not malicious. Feedback has been sent and anyone who feels the same way should send more. Either by the usual feedback link on the Takeout page or the Feedback form at.
https://support.google.com/accounts/contact/takeout_feedback
There's an associated problem with inviting or adding existing members to the group on the new platform. Takeout.G+Communities does provide VCF data for the members. But it's minimal and not terribly useful, consisting of name, G+ profile URL and that's about it. It's going to be important to set up the new community and announce it in the G+ community while people are still around and G+ is still active to give people time to find the new location and join again.
This is beginning to really annoy me and so I I find myself compelled to argue back.
Here's what I wrote last time this came up in
https://plus.google.com/u/0/102282558639672545743/posts/icWeMWvj8hr
Technically, the G+ API let's you get access to any post and it's comments that you can view from code. That's the technique used in the open source tool linked above. You do have to jump through hoops to get a list of the posts in a community in a form that's useable by the API. So Google does provide tools to save the content. They just don't wrap it into an easy to use Community Takeout download for owners/moderators.
Procedurally, I've seen this work with no problems on other platforms. There's usually a public debate within the group/forum/community about where to go. That ends up with an announcement that the content is going to be moved asking for objections. And then if there are none, you just go and do it.
The request in the OP is a FAQ. There is very clearly a non-trivial need and desire to archive and migrate the content in some active communities. And that need and desire is not malicious. Feedback has been sent and anyone who feels the same way should send more. Either by the usual feedback link on the Takeout page or the Feedback form at.
https://support.google.com/accounts/contact/takeout_feedback
There's an associated problem with inviting or adding existing members to the group on the new platform. Takeout.G+Communities does provide VCF data for the members. But it's minimal and not terribly useful, consisting of name, G+ profile URL and that's about it. It's going to be important to set up the new community and announce it in the G+ community while people are still around and G+ is still active to give people time to find the new location and join again.
There is a difference between making data publicly available (you have given consent to this before using Google plus), processing the data (you have given consent to this to Google and only Google before using Google plus) and allowing others to process any data other than your own (you haven't given Google permission to do so).. It may sound like splitting hairs, but it is the thin line that makes up data protection laws: prohibiting others from profiling everybody automatically without rendering services useless for everybody.
ReplyDeleteIn one of these exchanges with a Mod, I found a comment from him in that exchange. I gave it a +1. I then used Takeout to get the Activitylog and then posted publicly the "+1's on comments" file from Takeout. There at the top of the file was the full text of his comment, given to me by Google without his express permission. And I did the exercise to specifically point out that Google is already giving downloadable content of other people's posts. You can try and argue that this is a different situation to a Community Owner downloading community content from other people, but it's getting harder to justify.
ReplyDeleteWhile doing all this, I came across a post in G+Help from the same Mod a couple of years ago. He was arguing then in the same way that some function was impossible for Google to provide due to Privacy issues. Again it was something already available via the API. And shortly after, it appeared in Takeout.
What I don't get is the relentless negativity. Don't do that. You shouldn't be able to do that. Google can't do that. You shouldn't be asking for that. But then given the quality of the vast majority of posts in that community, perhaps I'm not surprised. It's a thankless task.
That's YOUR activity stream. You get the comment you collaborated on with your +1(ok'ish), but not all of the comments of posts you +1 a single comment on(big no-no). The keyword is data economy.
ReplyDeleteThe vagueness and threat of draconian punishment under GDPR create a very real incentive to err on the side of caution for any company with substantial revenue. At this point often the only winning strategy is not to play, maybe part of the reasons why G+ is being shut down in the first place.
ReplyDeleteOver the last few years, countless engineering hours have been spent across the online industry implementing such hair-splitting differentiations that make little sense except to an inner circle of GDPR-lawyers. And they probably don't know either, they are just being cautious...
Andre Naumann And yet I can do this via the API. Get the +1ed comment, find the ID of the parent post. Get it's content. Then get all the content of all the comments on that post. Providing it's public and I have visibility of the post in my browser, I can also get all the content via the API.
ReplyDeleteAnd this is the point. The S. D. Salyer, archiver at github.com - sdsalyer/gplus-archiver does nothing tricky or special or by scraping web pages. It uses the G+API. All this data is currently available using Google tools. Just not as a Takeout.
But then it's you breaking the GDPR rules, not Google and that's the point ;)
ReplyDeleteSo then let's take another related example. When I download the takeout of my posts, each post comes with all the content of all the activity on that post. Including the full content of every comment on that post. Without ever getting explicit permission from the commenter. I am the owner and moderator of the original post. And Google is giving me all the content from other people related to that original post. And it's just not a problem.
ReplyDeleteBTW. I consider the way GDPR and the Cookie issue have been written and implemented to be a ridiculous and over-bearing tax on internet usefulness. It winds me up every time I have to click though a popup dialog box saying "we do what everybody else does, please click to agree". It's stupid and pointless.
Your very own post and its comments are a collaborative effort, so you get access to it.
ReplyDeleteExporting a G+ community might include posts not written by you and not including a single comment by you. That's something you shouldn't have automated access to.
And if you read half the cookie popups carefully, you'll notice a slight difference in them.. some of them tell you that data about you will be stored(well, duh, it's the internet), some of them will tell you that data about you will be stored and shared with other people(no, thank you, I don't want other people to know that I just bought $WEIRD_SEX_TOY, it's bad enough that the web shop owner knows, but that can't be helped).
Or somewhat more extreme, but real example: I can't get insurance against occupational disability, because I've got a mild case of Asperger's. It often leads to depression and that a risk factor for insurance companies. Some doctor said so and that's why I can't get the insurance policy.
I wouldn't want to be denied that insurance, because I occasionally googled for "depression" or "Asperger's" and that's what the GDPR laws try to protect against.
The processes are still in their infancy and there's more panic than there should be, but the main idea is, in my opinion, still a good one.
Julian Bond I'm not sure that highlighting inconsistencies in Google's position is a winning strategy. It seems to me a result with non-negligible probability is that the features of the current API being used to show inconsistencies will be removed.
ReplyDeleteKent Crispin Well the API will get turned off in Aug 2019 anyway. Probably. It has only changed very, very slowly over the years. Consensus is that there's nobody working on it right now. In the mean time, activities.get, comments.list are pretty core to the API. I suppose it might get turned off early.
ReplyDeletehttps://developers.google.com/+/web/api/rest/latest/comments/list
/follow
ReplyDelete/sub
ReplyDeleteI understand the thought process behind it, but it doesn't make it any less frustrating.
ReplyDeleteJulian Bond I agree with Andre Naumann that we don't want Google acting like Facebook with our content. However, Google+ Community Owners may have similar ownership of content as post owners. This interpretation would facilitate community owners migrating communities, hopefully using ActivityPub json.
ReplyDeleteDisclaimer IANAL.
The question is a thorny one. The G+H mods tend to take an exceedingly conservative stance on many issues, which makes some sense when you consider they're interfacing on an official Google channel, as volunteers, to a very large membership (by G+ standards), of widely ranging skills and in varying jurisdictions.
ReplyDeletePlus online copyright is a poopfest of itself.
The general case that you could make an argument for some X is almost always true. The question you want to ask is whether or not you want to be compelled to make that argument, in front of a court of law (criminal or civil, depending on circumstances), possibly facing risks to personal or business property.
There is, in the United States, the affirmative defence of Fair Use. That is not a get-out-of-gaol free card, but a possible defence against copyright infringement claims which can be argued, should the need arise, and with its own case-law and legal gloss concerning it.
Odds are probably good that for personal use, on your own servers, the casual reference of individual comments would go without consequence. And the practice of ripping off and quoting work in whole elsewhere certainly happens. It seems at times my most fervent readers are spambots of various stripes.
Maybe it should be reasonable to consider that noncommercial public posting of incidental comments is a general public resource and can be posted without prejudice so long as those terms are preserved. But that isn't the case.
The small or individual case can likely get away with this. But there are risks.
Alternatives are to expressly require comments to be openly licensed -- Creative Commons or similar. This is the case at sites such as Wikipedia.
And a feature such as this might also be worth considering for future alternatives.
/follow
ReplyDeleteAlso, backing Julian Bond's comment, clearly, Google can technically export material and do, regardless of copyright claim.
ReplyDeleteExports for personal use would in general fall clear of copyright restrictions, as well.
That element of the objection raised by G+H mods is ... curious.
Edward Morbius - because the main issue is not copyright, but user data protection regulations.
ReplyDeleteBernhard Suter I'm not following.
ReplyDeleteParticularly for public information.
Edward Morbius - in a narrow interpretation (which most big-company lawyers might lean towards right now), the user data protection rules might also apply to data that is presented publicly in some sort.
ReplyDeleteBy uploading a post to a social media site, a reasonable person might expect to have consented to the narrow use-case of displaying that information on that site under a well defined mutual agreement and if the user changes their mind and edits or deletes the post, it should not longer be shown (erasure, revocation of consent). Giving that data in a permanent, non-revokable machine readable form to any 3rd party (without explicit consent of the user) could be seen as questionable.
Bernhard Suter Thanks. Yes, that's a possible interpretation.
ReplyDeleteGetting a solid GDPR read of risks is something I'd like to see us build out at #PlexodusWiki.
11 days till G+Sunset and Owners/Moderators were given Takeout.G+Communities for public communities. Except it was the posts only and not their comments. And you can't get the comments from the post even with the attached ActivityID because the API has been closed down.
ReplyDeleteThanks Google! And thanks Mods for arguing against this for so long. :(