Must have thoroughly documented encryption and security practices.
Must have a published audit from a reputable, independent third party.
All non-essential telemetry must be optional.
Must not collect more PII than is necessary for billing purposes.
Best-Case criteria:
Telemetry should be opt-in or not collected at all: +2
Should be open source and reasonably self-hostable: +2
Some people argued that open-source should not solely be the criteria into determining if a an option is recommended or not. I believe nobodyâs saying that. Open-source, though is clearly a plus for the only reason that it brings trust to that option. It is clearly a plus, but not the only criteria.
Based on that logic, if there are already 3-4 password managers that are on par with 1password feature wise, audit, wise, usability wise, security wise, etc. BUT onepassword is the only one not open-source, then IMO it should be removed.
Iâm not arguing this is the case though.
Edit: Iâm not suggesting a point system be in place. This was just for the sake of demonstration.
I guess the answer is yes. The criteria would remain unchanged unless one of the following is true:
The community wants to add something that is clearly so much worse than our recommendations (like Nordpass) that increasing the criteria is necessary to avoid confusion about why Nordpass isnât recommended.
One of the recommendations falls so far behind the others that it no longer makes sense to list.
The community wants to add something that is so much radically better than our existing recommendations that it necessitates delisting the existing tools.
None of these 3 three things really apply.
And again, the argument that you and @fria are making is that situation #2does apply, and Iâm countering this argument by saying that source code licensing doesnât create enough of a quality âdeltaâ between 1Password and the others to say that âit has fallen behind.â
Is this an accurate assessment of where weâre at?
I encourage everybody that hasnât done it yet to VOTE here on whether to require Source-First, Open-Source or keep allowing proprietary password manager.
So far, itâs a tie between source-first/open-source and proprietary
Then why do you have this criteria in other categories at all? Password management is as mission critical as it possibly gets and therefore it should have the strictest criteria, specially when it comes to anything âcloud-basedâ and/or that has to connect to the internet.
If thatâs the concern, then how about still mentioning it, but separating it from the other recommendations like what you do with e-mail providers? I understand the reason for Tuta being the â3rd optionâ is different, but I believe such a division would be warranted if you believe that FOSS is an improvement.
This is what I think a lot of users find strange. Obsidian could in many ways be considered the â1Passwordâ of digital notebooks, yet it is not recommended due to being closed-source despite being able to be used fully offline (with lower stakes in general), unlike 1Password.
Am I the only one who sees an obvious issue with other voters knowing what the entire PG team voted for? It might just be me.
Another issue is that some voters love 1Passwordâs UI and UX and are clearly biased against requiring open-source because their favorite password manager will get removed, but I disgress.
I am unaware of source-available password manager. I think this is unlikely. Source-available will be something you canât even build yourself (I think?).
So if you are OK with it, please vote for source-first .
As Proton Pass catches up with 1 Password, things might change, so thatâs a silver of hope.
Just because something doesnât exist now doesnât mean it wonât in the future. I find minimal privacy advantages in source-first or open-source compared to source-available, other than the increased chance that more people have reviewed the code. That is my personal opinion.
Ultimately, my vote supports changing the current criteria, but I have a different opinion on how extensive those changes should be.
Regardless, I donât believe this poll will have a major impact on the outcome of our discussion, even though I had hoped it would. The staff seems stuck in their decision, despite what most of their community is expressing.
Privacy Guides gets like 8,000 unique visitors every day, and this forum has over 3,000 members. One poll which has 34 votes and is split between 41% / 49% is unfortunately not the clear consensus you are imagining. On the contrary it proves that we canât proceed too quickly here.
If only it were that easy!
Much like Wikipedia, Privacy Guides is not a democracy, rather we make our decisions based on discussion and consensus. A poll is not a discussion, and this one frankly has hindered this discussion a bit as we are now all talking about the poll itself rather than the merits and drawbacks of this proposal. The polarization here as a result of the poll has really only slowed down the possibility of us making this change.
Do not worry, because we are considering the opinions shared here in this thread, in that poll, and across the wider community as we carefully chart out the best course of action here
My previous statement was saying that not requiring open-source in this category was the previous consensus, which is just an objective fact that it was. Now we are seeking a new consensus, which is fine, but it will take as long as it takes.
It is much preferable for us all to accept that it is not the end of the world for us to leave a less-than-perfect criteria on the website â with the understanding that Privacy Guides is always gradually improving and changes are being discussed and considered â than to try and fight for and push forward some immediate change for a particular preferred version of the criteria ASAP.
This is probably an off-topic discussion we could take elsewhere, but I do see PGâs âgovernance and consensus structureâ as similar to Wikipedia personally, and at least that seems to be the direction things have been heading.
This is also off-topic, but yes I would take our statistics with a big grain of salt. People sharing IPs or using Tor might appear as the same person to our (primitive) stats lowering the count, and on the other hand people switching IPs/circuits frequently may appear as multiple visitors increasing the count. So I think it is probably a wash basically, but who knows for sure.
Our pageviews are probably the most accurate statistic we have, which hovers around 35-40K/day.
Sorry, but that doesnât make any sense. The ratio between votes and total visitors here is irrelevant. There is always only a minority which actively participate, on any subject - so following your argument we couldnât take any decision cause too few people participated in it.
Sorry, but PG often doesnât make decisions based on consensus. When you or the team wanted something - like including Mull- you do itdespite huge backlash.
Plus, achieving consensus would require more tools, and a forum thread is not the best way to do this.
Some recent comments about that could be moved there.
Also, I closed the poll. I think I agree with Jonah, that while it shows a small majority supporting open, we might need to wait for a stronger consensus - even as I personally support requiring open-source, and believes there is a double-standard going on (see Mull (Android Browser) + Criteria Change) where team members and esp.
What is your deal with the 1Password? I am not paid by 1Password but I strongly defend them, even though I find their support and social media practices bad.
They are frequently audited, more frequently than any other password managers in the market, and that is not good enough for you?
I suggest you to write your own password manager, make it FOSS, and letâs see what happens.
You are speaking like FOSS is the next big thing, no, it is not. We, or I, regular people trust the audit companies, bug bounties and how companies react to the findings, becase I donât know how to code and how to properly check a code for a vulnerability or enhance it, like the majority of the world population.
You donât need to be paid by 1Password, just the fact that youâre using it is enough.
So what? How does that matter? 1Password has one annual pentest per year, which is next to useless.
Who knows if the auditing company is competent enough?
Who knows if the exact people from that company that did pentesting are competent enough?
Even if the auditing company and the people who did the audit were competent, theyâre unfamiliar with the code because they arenât the ones working on it.
Even if the audit was done perfectly and flawlessly in every way, 1Password is only doing audits annually, and the vulnerability can be introduced in the next update right after the audit.
There is literally a year gap in which the 1Password developers, maliciously or not, could introduce a vulnerability, and if it was in fact malicious, get rid of it before the next audit.
Youâre just a user of 1Password who likes how it looks and feels and who is completely used to it, thatâs why you defend it so much, but Privacy Guides isnât about âlooksâ and âfeels,â at least I hope so.
Good, I will make a password manager, get it audited to get the trust of regular people, and then introduce 10 backdoors until the next annual audit.
.