Please do not Recommend Services Using Google Analytics

[melvincarvalho]

It was pointed out by @csarven that the gitter chat room shares user data with Google Analytics

Google Analytics is generally considered to be privacy invasive and is not aligned with the core values of the Solid project

Chat is integral to the participation and evolution of the project. Users coming to the the project for the first time, and having strong views about user data, may later learn that they had been directed to services that did not respect their privacy, and be understandably upset

This being the case, I remain of the view that repositories under the solid org, do not recommend services which use Google Analytics and share user data with third parties, especially when no alternative is offered

[elf-pavlik]

people can use bridge from other services, i know couple of people who use matrix and join gitter from matrix over bridge. https://matrix.org/bridges/#gitter

[csarven]

To be accurate, I did not say or know for a fact that "gitter chat room shares user data with Google Analytics". What I said was "Gitter includes Google Analytics JavaScript." - https://gitter.im/solid/chat?at=5e468cd2292ff243d3c8c77d . It could be anonymised data - I'm not saying that's okay or not either. If anyone is familiar with the GA codes or whatever, it may be useful to know what kind of data is being transmitted by default (ie. JavaScript enabled and GA scripts are not blocked).

I've suggested browser extensions here to help users as (an interim) solution: https://gitter.im/solid/chat?at=5e46b6e025f1d250fed7ae13

Some more thoughts and ways forward for the community here:

• https://gitter.im/solid/chat?at=5e46cdb85f82805026d9a014
• https://gitter.im/solid/chat?at=5e47e7a025f1d250feda67a3

Proposed that Melvin (but anyone for that matter) may want to have a closer look at reasonable options, evaluate, document... and share their findings with the CG (in a call).. run it by the Director (as per process) etc.: https://gitter.im/solid/chat?at=5e47f431292ff243d3cc25c5

I don't have anything in particular to add beyond that at this time. Good to have this issue in our radar but for anything to happen, we need actual data.

Aside: I don't particularly see an issue with having multiple ways (channels, platforms, protocols.. ) to participate in the Solid community either.

[jonassmedegaard]

Frankly I am surprised this is news at all. I avoid gitter (same as I avoid the commercial video chat offerings for the weelḱly meetings), and cannot imagine I am the only one.

Whether I and the likes of me are any relevant for the Solid community to care about I cannot say.

[melvincarvalho]

Frankly I am surprised this is news at all. I avoid gitter (same as I avoid the commercial video chat offerings for the weelḱly meetings), and cannot imagine I am the only one.

Whether I and the likes of me are any relevant for the Solid community to care about I cannot say.

Couldn't agree more!

Solid should be for everyone

[jonassmedegaard]

for the record (since the followup by @melvincarvalho might give a different impression) I do feel welcome to join both text chat and voice chat, and I do not condemn your choice of tools as fundamentally bad.

My point is that I choose to avoid commercially tracked services and to prioritize serviced based on Free software and open standards over closed alternatives, and when you choose differently then I won't hang out with you - in case you find it relevant to have me aboard.

Reason I point it out is that I realize that my absence and its correlation with your choices of tools might not be noticed unless I point it out explicitly.

[elf-pavlik]

Gitter has its source code open: https://gitlab.com/gitlab-org/gitter
As for voice calls, I think some panels use https://meet.jit.si/ which seems working pretty well. I wouldn't mind adopting it for more panels and start moving away from Zoom and GoToMeeting.

[melvincarvalho]

I believe I tracked down the code used in gitter which uses GA.

https://gitlab.com/gitlab-org/gitter/webapp/-/blame/f93f53f46cc27a1de92e7bc8b638752e5d3063e3/public/js/utils/tracking.js#L24

It appears to have been added 1 year ago, ie a few years after the project started using gitter

@timbl has suggested moving bit by bit from gitter to long chat

One possiility is to copy everything into Solid like https://timbl.com/timbl/Public/Archive/solid/chat/ and then just all one day start talking on the solid version instead of the gitter version

which seems a good long-term suggestion

The W3C Community Group also has an IRC room. I have personally started using this more, and the matrix bridge that @elf-pavlik suggested

I looked briefly at Zoom and that also appears to also share data with a third party via GA. I think it's wrong, to both claim having strong views about privacy, while recommending apps that spy on you

@jonassmedegaard I very much enjoy interacting with you. The project benefits immensely from a diversity of views, such as yours. In a general sense, I am a believer in offering choices that encourage inclusion. This issue is specifically directed towards those, that care a lot about privacy, which I think is the prevailing mainstream view, in this project

[mediaprophet]

My 2 cents

I have a different problem with gitter, which all link back to 'privacy' (or moreover 'dignity').

1. Precedent definition of rights/responsibilities in seeking protect 'privacy'.
   Under AU Privacy & FOI law there's an array of rights/responsibility mechanisms surrounding a series of principles whereby making corrections is defined by precedent, in relation to 'privacy', to be important. I am sure other examples exist with respect to defining frameworks surrounding the concept of 'privacy' and/or 'dignity' preserving.

2. Sense making & Accountability
   There's an ability to change text in a manner that acts to 'distort' sense-making about 'causality'; and, there are complex matters which in the interests of privacy, dignity and related factors should not be made public; which can in-turn be used as an attack-vector, and, the identifiers aren't necessarily associated to a 'real person'; alongside other factors relating to https://en.wikipedia.org/wiki/Information_Operations_(United_States) exploits, et.al.

3. Helping to make infrastructure to project 'human dignity' vs. web slavery for commercial entities
   There is also a distinction between implementers and a standard effort, such has traditionally been carried out under the stewardship of W3C (inc. Community Groups). Whilst the work carried out by people in the interests of making free and open standards protected by vast patent-pool holdings of legal personalities across the world is done by some without being paid or provided direct remunerative benefits for doing so; many are willing to do so, as the consequence of the act of doing so is considered to be important for the future of our world and core tenants of our society, biosphere. It is doubtful that people would want to make the same commitment to an effort that may be for the commercial benefit of a particular commercial entity, located in some jurisdiction somewhere in the world. Whilst this is a complex issue, i think its important to have a distinction; thats defensible on a practical basis.

4. in terms of a universal 'community of practice' for the web - W3C is in-effect 'IT'.
   Last year, i set-up a zullip server which i note also has LDAP auth alongside other features - which in particular associate to the way conversations are structured are topic (and/or perhaps via API URI) based threads, which i found useful from a productivity point of view, although it was nuanced from traditional 'slack like' methods.

Zulip also supports video conferencing, et.al. via jitsi which is also open source.

Perhaps W3C might be able to get something working on a trial basis, where its connected to the CG Contributor agreements (etc) and that AUTH is facilitated via that LDAP server.

[jonassmedegaard]

@mediaprophet I recommend to file a separate issue for separate (even though possibly related) issues.

(related to that, I realize that I arguably made same mistake myself in my contributions to this issue - communication is tricky)

[mediaprophet]

It was my understanding the issue was about privacy. What happens when an app parses user data with something like https://search.google.com/structured-data/testing-tool ? Privacy moreover had a legal definition.

…

On Wed., 19 Feb. 2020, 10:13 pm Jonas Smedegaard, ***@***.***> wrote: @mediaprophet <https://github.com/mediaprophet> I recommend to file a separate issue for separate (even though possibly related) issues. (related to that, I realize that I arguably made same mistake myself in my contributions to this issue - communication is tricky) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#270?email_source=notifications&email_token=AIKBW2M6NQNEMFKEXHV425DRDUH7HA5CNFSM4KV3A5C2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMHLXOY#issuecomment-588168123>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIKBW2IBMVNR4VVBON7CH3DRDUH7HANCNFSM4KV3A5CQ> .

[jonassmedegaard]

Quoting mediaprophet (2020-02-19 12:25:10)

It was my understanding the issue was about privacy.

Issue is this: "Please do not Recommend Services Using Google Analytics" Yes, that _relates_ to privacy and other things do too, but that does not mean that this issue is about me being naked in our video conferences, or other somewhat related topics. - Jonas

…

-- * Jonas Smedegaard - idealist & Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private

[mediaprophet]

"that does not mean that this issue is about me being naked in our video conferences"

…
-- * Jonas Smedegaard - idealist & Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private

its the first i've heard of such a behaviour, and as that's the first i've heard of any such idea temporally, i certainly wasn't making any attempt to attend to that type of use-case in my response.

[ghost]

Interesting, this thread now being 2 yrs old, i wonder what tools others have found for thier use case and threat model to prevent unwanted tracking. Before i speak more to tools i use,,and balance between security and privacy , out of full respect and loyalty to Sir Tim and not to innundate Solid's mission... I look forward to hear what, if anything you here on this thread maybe doing now, 2 years later regarding this issue?

[mediaprophet]

Wrote this in 2018: https://www.webizen.net.au/about/executive-summary/preserving-the-freedom-to-think/ Now I'm seeing stuff like this, https://twitter.com/mazemoore/status/1522620554011938816 As there's an election in Aus being carried out on foreign platforms. IMO: problem, due to moral poverty, has become far worse, not better... There's no #RealityCheckTech and I fear the consequences of making that technology is increasingly opposed to the interests of those who've been defined as leaders, whilst others have been excluded as work done without income, turned into a commercial opportunity for others. The underlying problem is ideology & commitment to rules, particularly rule of law. If no commitment to rule of law, then it doesn't really matter what the rules are, they can always be changed (as can the data / evidence), which isn't what I wanted for VerifiableClaims - but it's what's happened. Pity. FWIW: rather than data collars, DIDs should support "web3Ontologies" (permissive commons) & other alternatives for discovery and privacy/dignity support whilst operating AI/Software agents, associated to human agency. But I'm not sure my ideologies around epistemological identity theory is supported, particularly in areas where natural persons were intended to be the primary beneficiaries... #Realitychecktech was about telling the truth, not about rewriting history to distort reality / make defunct rule of law, etc. So, simple answer, checkout DIDs. I think that's the present day preferred ideology, etc. Not sure how the solution supports lawful remedy or reputation... Drew the attached diagram about reputation systems, trying to describe the differences in ideology between systems intended to impactfully execute consequences based upon claims, vs. those that are designed to support judgements in Courts of law. There's significant differences in the effect systems have based on the ideologies of implemented designs. Which links to an issue about privacy vs. dignity, which matters less to incorporated legal personalities, than it does to natural persons... https://twitter.com/DemocracyAus/status/1521721199742832640 https://en.wikipedia.org/wiki/Giant_Global_Graph I don't know how democracies could be made to work if there was a super user on a platform like akamai that had ACL control over everything for a price or fee. But trust is more like a graph of behaviours, I'm not sure how much effort has gone into related real world considerations. Cheers. Timo.

…

On Sat, 7 May 2022, 12:21 am MaxSolidEra, ***@***.***> wrote: Interesting, this thread now being 2 yrs old, i wonder what tools others have found for thier use case and threat model to prevent unwanted tracking. Before i speak more to tools i use,,and balance between security and privacy , out of full respect and loyalty to Sir Tim and not to innundate Solid's mission... I look forward to hear what, if anything you here on this thread maybe doing now, 2 years later regarding this issue? — Reply to this email directly, view it on GitHub <#270 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIKBW2N5P6LYNYFBRT6PKPTVIUTGDANCNFSM4KV3A5CQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>