For me that’s been since August 2022. For my partner since December 2022, on an HP laptop, and later, June of 2024, a Starlabs Starlite 2 in 1.
Then again, you’re known not to abuse your system. Any time the topic arises, the general state of a user’s environment isn’t generally factored into the conversation.
Those with a general disregard (or ignorance) of system maintenance requirements may experience more perceived issues with pamac-manager than there might actually be.
As i said previously
I suspect the issue, is not the pamac GUI, but some thing done or left undone by those users who end up having issues.
Some how, I suspect, if they had started off with pamac-cli, they might still have had issues.
Snap!
![]()
Reflecting on my earlier comment, I’m tempted to qualify it by adding that issues I have with pamac-manager tend to be on the rare occasion I use it; which in turn, is usually after reading that someone is experiencing difficulty.
Otherwise, I use Octopi if I’m too lazy to type.
Stable update polls for 2025 are a long way from that:
| Date | No-Issue | Total Votes |
|---|---|---|
| 2025-04-12 | 90% | 302 |
| 2025-03-30 | 82% | 281 |
| 2025-03-24 | 83% | 267 |
| 2025-03-18 | 89% | 260 |
| 2025-03-05 | 81% | 295 |
| 2025-02-16 | 80% | 270 |
| 2025-02-04 | 84% | 331 |
| 2024-12-16 | 93% | 270 |
Polls for 2024 shows highest percentage result of 95% and a few with 94%
Lowest percentage was 61% (2024-05-13) with 408 total votes (61% of 408 = 249)
It would be interesting, if we do a segmentation by DE in the poll.
Then, we can see if there is any DE that has lower score than other DEs.
The number of users voting is probably (hopefully!) very small compared to the total number of users. I doubt any useful conclusions can be derived from the poll. It is fun, but i think that’s all it is.
True — and people tend to be less likely to vote when there are no issues.
Probably. Although there is no evidence anywhere.
Incidentally.
Similarly, I think many people will decide to just leave Manjaro.
There are many places to go. That is also a healthy decision.
I think most people reading this are adults who are in business, so I think they understand that in today’s commodity market, it is sometimes the actions of the silent majority, not the loud minority who are in sync, that hold the casting vote.
Justice does not always win, and whether that justice is correct in the first place is a completely different discussion. To put it mildly, no one knows.
I can confirm this as i am responsible for our global helpdesk in my company with approx 250k contacts every year
We are sending out a survey invite on every ticket we are closing. However the return rate is currently at 8%
So the other 92% are either satisfied, too busy or tired of the helpdesk.
Customers are not giving positive feedback as they say “well the helpdesk works as we are paying for. Why should i give feedback?”
Slightly OT: I always respond to Telstra requests for feedback, as invariably one of the questions is ‘would you recommend Telstra to others’ my reply is invariably ‘No’ but that doesn’t seem to have improved their service.’
If there’s one thing guaranteed to cause me great annoyance it’s the constant stream of emails I get from so many companies nagging me to rate their service just because I bought something. My reaction is always to give the lowest possible rating with a comment along the lines of “Don’t buy anything here unless you really love getting nagged for ratings”.
Customer pays, helpdesk provides support.
A balanced cycle.
Company asks feedback, provides no return.
No balance, no gains for customer.
Customer has no knowledge on if the feedback will impact the company’s policy or service, if any.
Just another paper practice? Most likely.
Worth the time? Nah.
I agree on this, but i am talking about internal IT support for our clients in the company. This is given to incidents only, not orders or anything else.
So the lower the amount of calls i need to the helpdesk, the less mails i receive ![]()
But now back to topic.
Manjaro does not publish information about the total number of systems updating but I would estimate that 95%-99% of systems are not represented in the poll
non-voting users probably have no issue to report or an issue was already reported and resolved
If users experience an issue and do not contact forum for support there is nothing to be done
I believe this is not technically feasible.
And that would be a good thing.
Ultimately, if a support forum has very few support cases lodged, it either means:
- users have abandoned the forum,
- the product is so flawless that nobody has an issue.
The traditional response (having 500 kg of horse manure delivered to the head office lobby) isn’t as practical as it may once have been; partly thanks to globalisation; so a donkey rating may be the next best alternative.
Surveys can give a good opportunity to raise awareness of issues that may affect any given userbase; whether the information might actually be used for much more than landfill remains unclear, despite corporate SOB.
Perhaps this has something to do with the fact that a company’s FAQ also never actually contains any frequently asked questions, but rather rarely-ever asked questions that the company’s marketing and PR departments would be all too happy to answer because they like patting themselves on the back.
Even if only subconsciously, people are generally aware of the dishonest intentions behind such inquiries — and especially so if the inquiry is presented to them by an independent marketing firm — and that’s why they’re generally not inclined to partake in such surveys.
This reminds me of the survey sent out when they removed a local viaduct — the survey questions were LOADED in their favour. I should have sent it back with a note explaining this, and my dissatisfaction with the decisions made. (How, exactly, is taking a viaduct out and carving up a popular walking and cycling route with a level crossing, an improvement?)
In our case, it’s a pity mdd didn’t go down too well, because that was FAR from being dishonest; quite the opposite, in fact. That way we’d have a much better picture.
Yes, I agree. But the hateful Redditters and the Arch-btw’ers deliberately sabotaged that attempt.
Apparently it is always easier to destroy than to (help) build. ![]()