News: 1689015605

  ARM Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life (Terry Pratchett, Jingo)

Fedora Project mulls 'privacy preserving' usage telemetry

(2023/07/10)


The Fedora Project is considering a proposal to introduce some limited usage telemetry in a future release. Predictably, quite a few users are not delighted with this development.

The suggested "privacy preserving" [1]change would introduce some very limited telemetry into Fedora 40 next year, and as far as we can tell, this would only apply to the default GNOME desktop edition. The accompanying discussion is, as you might expect, already quite spirited.

The proposal actually seems fairly modest. It would be fully anonymized, not collect any personally identifiable information, not collect search queries, and so on. Last time we looked, the project hadn't got as far as defining exactly which stats would be collected.

[2]

The proposal comes out from the Red Hat Display Systems Team, the part of the organization that maintains the desktop operating system. In case the name isn't familiar, these are the folks who recently decided to [3]stop maintaining the distro's LibreOffice packages.

[4]

[5]

They propose to use the usage metrics gathering and processing tools developed for the Endless OS distro; we [6]looked at Endless OS version 5 at the start of the year. One of the Endless project's developers gives some details about its info-gathering tools, collectively known as [7]Azafea , in a [8]blog post . Additionally, the Fedora change proposal clearly specifies:

We do not plan to deploy the eos-phone-home component.

The [9]eos-phone-home tool, amusingly, is a variant of an Ubuntu tool from some years back called [10]canonical-poke , as described in [11]this old blog post, and [12]discussed on Stack Overflow soon afterwards. The point here is not that any of this derives from Ubuntu efforts or anything of the kind; the point is that things like this are nothing new.

[13]Forester delivers bare metal remote provisioning to Fedora

[14]Rocky Linux claims to have found 'path forward' from CentOS source purge

[15]Red Hat releases RHEL 9.2 to customers, with buffet of rebuilds for the rest of us

[16]Red Hat promises AI trained on 'curated' and 'domain-specific' data

Back in 2010, the GNOME project conducted a [17]survey , and [18]Ubuntu got caught up in the ensuing controversy. Even Debian has [19]Popcon , its software-package popularity contest. Previous related Fedora efforts, mentioned in the proposal, were the [20]Smolt tool and [21]DNF Better Counting .

No development team can do everything that it would like to do. Even volunteer-driven community projects have to find some ways to prioritize their efforts: to work out which parts of their systems are being used the most, and which the least. The Fedora proposal is open about the kind of information that the team is looking for:

One of the main goals of metrics collection is to analyze whether Red Hat is achieving its goal to make Fedora Workstation the premier developer platform for cloud software development. Accordingly, we want to know things like which IDEs are most popular among our users, and which runtimes are used to create containers using [22]Toolbx .

The proposal specifically addresses the vexed issue of opt-in versus opt-out, as well as the separation of data collection and data upload, and other telemetry related FAQs – something which appears to have been missed by many of the people commenting upon this online in various forums. Perhaps the key point is that it will be on by default, because as the proposal says, the developers feel that they must:

… ensure the system is opt-out, not opt-in. This is essential because we know that opt-in metrics are not very useful. Few users would opt in, and these users would not be representative of Fedora users as a whole. We are not interested in opt-in metrics.

The Reg FOSS Desk suspects that many of the reactions are essentially knee-jerk ones, and that many of the people who are already objecting have not read the proposal in detail. 'Twas ever thus with internet bun fights.

Every distro has attempted things like this before. It's really no biggie, however vocal some of the complaints. The Mozilla Firefox market share [23]is now down to about 3 percent, and yet we are willing to bet that many of you will be reading this article in some variant of Google Chrome —- despite Google's well known data-collection habits.

Android does a great deal of phoning home, which is [24]why Murena /e/ OS exists , and yet billions of people use Android devices. Unpopular as it was, we still think [25]Scott McNealy was right in 1999: "You have zero privacy anyway - get over it." ®

Get our [26]Tech Resources



[1] https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/55H3DT5CCL73HLMQJ6DK63KCAHZWO7SX/

[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2ZKx-hoo5-qOqWS2IkcWHWAAAABY&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0

[3] https://www.theregister.com/2023/06/07/red_hat_drops_libreoffice/

[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZKx-hoo5-qOqWS2IkcWHWAAAABY&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/front&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33ZKx-hoo5-qOqWS2IkcWHWAAAABY&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[6] https://www.theregister.com/2023/01/12/endless_os_5/

[7] https://azafea.readthedocs.io/en/latest/

[8] https://blogs.gnome.org/wjjt/2023/07/05/endless-oss-privacy-preserving-metrics-system/

[9] https://github.com/endlessm/eos-phone-home

[10] https://launchpad.net/canonical-poke

[11] http://theravingrick.blogspot.com/2010/08/can-we-count-users-without-uniquely.html

[12] https://askubuntu.com/questions/14351/what-is-the-canonical-census-package

[13] https://www.theregister.com/2023/06/22/forester_bare_metal_remote_fedora/

[14] https://www.theregister.com/2023/06/28/rocky_linux_rhel_ripples/

[15] https://www.theregister.com/2023/05/19/rhel_92/

[16] https://www.theregister.com/2023/05/26/red_hat_ai/

[17] https://blogs.gnome.org/bolsh/2010/07/28/gnome-census/

[18] https://www.theregister.com/2010/09/14/shuttleworth_gnome_contributions/

[19] https://popcon.debian.org/

[20] https://en.wikipedia.org/wiki/Smolt_(Linux)

[21] https://fedoraproject.org/wiki/Changes/DNF_Better_Counting

[22] https://containertoolbx.org/

[23] https://www.theregister.com/2023/02/14/thunderbird_115_supernova_complete_overhaul/

[24] https://www.theregister.com/2022/06/02/murena_e_foundation_phone_test/

[25] https://www.theregister.com/2006/02/21/protecting_consumer_privacy/

[26] https://whitepapers.theregister.com/



Stats please

Sandtitz

"The Mozilla Firefox market share is now down to about 3 percent, and yet we are willing to bet that many of you will be reading this article in some variant of Google Chrome —- despite Google's well known data-collection habits."

No need to bet - perhaps El Reg could just share the browser stats.

Re: Stats please

PeeKay

Google Chrome varient? Yes.

Data collecting? No.

Vivaldi Browser. https://vivaldi.com/privacy/browser/

Re: Stats please

diodesign

Quick look from 2023 YTD, number of unique readers in that time:

Chrome = 52%, Safari = 24%, Firefox and Edge = 6% each. Rest a mix of mobile view on Android, in-app Safari, Opera, etc.

StatCounter has, for June 2023, global browser usage: Chrome = 63%, Safari = 21%, Edge = 5%, Opera = 3%, and Firefox = 3%.

C.

Re: Stats please

alain williams

That is why I use Firefox and not Google Chrome.

OhForF'

We are not interested in opt-in metrics.

So will you violate the GDPR or just not collect any data in the EU region?

I'll have to read up what they claim they want to use the telemetry for. Wasn't the Gnome team's usual reaction to feedback to ignore it as they know better than the users what is needed?

demon driver

I get that GDPR requires opt-in for collecting personal data. But if I understood this correctly, Fedora does not want to collect *any* personal data. What they want to collect is impersonal data. In that case, my impression is that GDPR does not even apply. Or do I make a mistake there somewhere?

icesenshi

To think that people actually get paid to write this drivel. Seriously, who do you work for, Microsoft, the king of telemetry? Telling the readers to get over it is just tone deaf, but I guess I shouldn't expect anything more from a hack.

usbac

I'm sorry, but I expected something better from the Reg. This give up, and get over it attitude is why we have very little privacy. The register is the last place I would have expected such bullshit to be spewed. The Register is becoming just another bland tech news site. The Americanization of The Register is not doing it any favors (and I say this as an American).

The fact is, opt-out telemetry is NOT acceptable in any software from any vendor, at any time, and especially from open source.

If developers think they must have it, they need to look into another line of work. We had computers and software that worked (in most cases much better than it does now) for four decades before there was any mechanism to relay telemetry back to the developers.

Familiarity breeds attempt.