Feature portal update

Announcements
1

Dear Checkmk users,

More than a year ago, we launched the feature portal as the central place to collect your ideas for improving Checkmk.

Since then, we have implemented 53 suggestions, including

Since the Checkmk 2.2 release, we have also continued to implement suggestions such as

Many more, especially on the user experience side, are planned for the coming months.

During our Checkmk Conference and many discussions with you, our users, we received a lot of feedback about the feature portal and your expectations of it. As a result, we have changed the default sorting option from “most voted first” to “trending suggestions first”, an option recently implemented by the tool vendor.

Containing more than 800 ideas, the usability of the feature portal has become a major challenge for many users. We recently introduced tagging which makes it much easier to navigate through all these suggestions. However, at the current rate at which we are receiving new ideas, it is only a matter of time before the existing setup is no longer sufficient to manage the flow of input, at which point the feature portal would become unusable.

We will therefore introduce regular portal maintenance from October 2023 on. As part of this, we will be closing suggestions that have no traction:

  • Suggestions with low user interest: created more than 1 year ago with 5 votes or less
  • Suggestions with no momentum: no votes in the last 6 months

As part of these changes, we have also given the portal a name that better reflects its current purpose and limitations: an Idea Portal.

In order to continually improve Checkmk, we value and need your ideas. They provide valuable input to our roadmap discussions and allow us to open a dialog with you for deeper discussions, especially during development.

The latter part in particular has been a great help to us. For example, for the top 10 hosts dashlet suggestion, we interviewed more than 10 users in detail and learned a lot about the underlying problems, use cases and expectations behind such a broad topic.

There are many further improvements we can think of that would also help us get the most out of the portal. The current software has a number of limitations that prevent us from doing this, so please bear with us until we can do a complete overhaul of the platform. The priority for the product team right now is to help build Checkmk 2.3 and plan for upcoming features.

If you would like to discuss these changes and future plans, please join the upcoming community call on the 17th of October, at 15:00 Berlin time. We will provide a brief overview of the changes and we are looking forward to hearing your feedback as well as your questions: Zoom Link |
Checkmk Community Call.ics (939 Bytes)

Thank you for your patience and continued support,
Your Checkmk team

8 Likes
2

Hello Checkmk Team

Having all of our ideas deleted by the definition of “having no traction” is not acceptable at all!
The rating is much too shallow.
It is not possible for us customers to look at all the ideas and rate them.
The ideas from us customers are relevant to the software we administer every day. That’s why we have entered them and thus collected them for development as an aid to how the product can evolve, because we have some problems or just see potential for improvement.

Please find another way to administrate such ideas, so that the ideas can flow into the software of Checkmk and can also be seen in the portal.
Deleting is not an option, so we lose the overview and enter the same ideas again and again, or what is more fatal we don’t have the information about them anymore.
Because for us customers are also ideas and thoughts that we share here with the Checkmk development and other customers.
If necessary, a certain label is good for this, according to which one can then filter or filter out.

But please, please do not delete!

Best regards
JPH

3

They are not deleted. They will get the status “Not planned”. Which means they can still get upvoted on and that they can still be found and commented on (e.g. Adding host labels via local scripts)

Happy to discuss this in the community call and hope to see you there.

4

Hello Martin

Thank you for the information of your approach, that the ideas are not deleted.

Best regards
JP

5

A quick reminder: we will have a call to discuss the updates in 40 minutes. Here is the link: Launch Meeting - Zoom

6

Thank you again to everyone who joined the call today!

Here is the presentation:
Community call: Checkmk Ideas Portal.pdf (843.7 KB)

1 Like
7

Did a quick look at the pdf.

What is the difference between your interpretation of implemented and released?

I assume


  • implemented: code ready but not yet connected to a (public) version
  • released: code ready and connected to a (public) version

Correct?

8

Yes. It was explained in the call.
implemented: merged in master branch, thus development is done
released: available in a stable branch

1 Like
9

As missed the call - anything different/new compared to the info mail?

10

Hi foobar, while we went a bit more into detail during the call, the most important topics have already been covered by the info mail. If you are interested in the topics discussed during the call, you might want to have a look at the slides attached to Sara’s comment above.

1 Like
11

Hello,

we introduced some new states, which we will start to use in the future and hope they will be helpful. You can find more about this here: Ideas Portal Guide | Checkmk

Happy monitoring!

12

Hi Martin,

Paying customers with large and/or complex environments like us are feeling in some way disadvantaged in the FR portal compared to the many Free Edition customers. This FRs usually get much less attention because the Free Edition customers do not face these issues and therefore do not vote for these FRs.

Another point is that older FRs are disadvantaged by the “New” and “Trending” filters. Customers who enter later do not see these. Only a few will make the effort and also look at older FRs.

How do you intend to reflect this in a visible way?

Regards, Lars

2 Likes
13

Hello Lars,
I understand your worries. Therefore we want to move to a different tool over time, which allows only voting for registered users. Then we also know which users voted for it and we know our customers with larger and more complex environments and can thus better evaluate the individual suggestions.
In the meantime, thus, use a real email address when you vote. When including suggestions into the roadmap, we look at the votes behind the suggestion as well. If we see that an idea has a lot of upvotes, especially from a lot of larger customers, we internally consider this with a ‘higher weighting’.

Older suggestions at the moment have the advantage, because at that time there were only a few suggestions which accumulated a lot of upvotes. Newer suggestions suffer under the sheer number of suggestions - something which we try to address with tagging and the clean-ups to get to a number which can be better handled by users. Still far from ideal, I admit. And sorry we closed quite some of your suggestions. But we are making an effort to get to a better situation at the moment (e.g. we discussed the top 25 suggestions today in product management and as a next step we plan to get buy-in for our proposals from the other teams as well).

2 Likes
14

I don’t want to make a rant post or anything like that, but if the most requested feature, by all features, during this entire period the portal have been open is to support Checkmk being installed on a Raspberry Pi (ARM)


Along the 10 most popular votes only one feature is planned. Along the 20 most popular votes only 6 are being planned.

Today I woke up to having 15-20 emails with “rejections”. One of them have already been implemented (but was rejected)

One request around security was rejected.

Spending valuable time creating feature request will at least from my side stop.

2 Likes
15

Hey Anders,
apologies, but the rejections are based on the clean-up rules, which we communicated with this thread. We know that on an individual basis, this is not great. But on the other hand, there was a very strong wish from users and from us, that we need to clean-up the long tail of suggestions.

Regarding the most popular suggestions:
When we looked at the top suggestions in early 2023, we selected quite a few ones which were in the top 20. As they were then marked as planned and most of them are also done, they don’t rack up any votes anymore. Thus, naturally the top voted features at a later time will not reflect this anymore.
The product management team today discussed exactly these ones and has made a proposal for each of them, if, when and how we want to do them. We now need to agree also as a company on that proposal.
We already communicated a clear statement on ARM.
Other like the Mobile App include a long-term 7-digit investment from our side, thus need to be carefully weighted and prioritized vs. other topics. Thus, we can not just make decisions lightly as most of the things in the top 20 are actually quite large topics, incl. features which are not developed by Checkmk (NagVis).

EDIT: Security-related topics, especially vulnerabilities fall might fall under Vulnerability disclosure policy | Checkmk and should be reported in that fashion. The feature portal is not the right place to be report security issues IMHO

16

As already partially pre-announced in the initial post: we will be changing the domain from features.checkmk.com to ideas.checkmk.com today. Expect a short downtime therefore.

17

Dear Sara and Martin and all CMK Users

With the announcement on the conference#9 you made it already clear and shocked quite some of us who invested lot of time over the past year into it, that for you its just a pool of ideas, and actually voting is just an indicator! :worried: Which sounded much different when you announced it in the first place and moved all pending customer feature requests from our support ticket system into this portal. At that point we still had little hopes our features will, at one certain point, find a way into CMK. - First announcement by your PM was “the ones with the highest votes will be implemented” (later adapted/changed already)

If I remember correct, on the conference#9 the result on the board was voted “priority list” over “idea portal”. So now its called “Ideas Portal”. Mhh again, might be not a good feeling for those who took the time and voted. :smirk:

This for example is not OK/fair for some reasons
For example, you implement features with less then two digit votes and one with 30 votes you will later drop? You can imagin how that feels to the dozens of feature requests proposed/written and voted by many of us users and now dropped?
And I would bet!, some of the by you picked feature requests would never reach 30 even after one year. They might not even get 1-2 votes per year.
Of course we all understand that there are features which are far more complex and time consuming then others, but so are there also some on the same, or even with a smaller
dev scale with higher votes, compared to the ones you picked with less votes.
As already mentioned here: Voicing your questions about Checkmk Development
One of the problems is, everything behind page 1 will only be found, IF

  • people active search for it
  • the search request matches the description
  • people are willing to spend a LOT of time on the portal to check for specific FRs
  • people on a regular base browse through the potal - and why should they if they only had 1-2 FRs they are interested in and voted for?

I would suggest at least combine it. No momentum in one year and less then 10 votes.

The 1year/5 votes been communicated well on the conference (last year already?) and in the forum long before - So quite fair for all feature portal users (regular and very seldom)
The new “rule” came without warning (or did I missed it?!) and I think not fair especially for those who voted and invested time. As for example @Anders mentioned - suddently all the work and motivation is gone.
As you can feel, its already related with the problem that for us users it does not feel fair or worth it to vote and invest time into it anymore (we had this discussion already before and now another setback for us? - not going into it again)

I will propose again the idea of introducing an internal “hackathon” where your devs can go though the list and try to solve as many of the feature requests as possible.
Long time ago something similiar was done for many years in a row in your company and
for us it felt really great because with the todays definition from CMK, at least half of been not Bugs, its been little thinks missing, which are today called features and we find plenty of this little things ind the feature portal. even if just 10 votes. And the feedback back in the day from your devs was also very good - as it was a fun gathering and even teambuilding as well.

Waiting for it :upside_down_face:

Good parts

  1. about the post - at least it was all communicated in a proper newsletter (lack of communication in the past, ‘only forum, only conference,
’ → well improved)

In the meantime, thus, use a real email address when you vote. When including suggestions into the roadmap, we look at the votes behind the suggestion as well. If we see that an idea has a lot of upvotes, especially from a lot of larger customers, we internally consider this with a ‘higher weighting’.

We’ve been contacted now a couple of times so thats some feedback where we can feel something is going on compared to the blackbox of feature requests before.

Just some feedback not a rant post! :hugs:

Cheers

18

Hey Daniel,
thanks for your feedback.

We offered a community call to discuss this and would have been open for your proposal regarding “no momentum in one year and less than 10 votes” then. Also, we send out an email with the proposed changes to everyone who voted or created a suggestion to react on this. We gave ample opportunity to propose or discuss this.
We considered also thoroughly what is the right limit, and the number seemed to be a fair number given that we need to reduce the number of suggestions. This was the clear wish from users on the conference AND in our discussions post-“Voicing your questions
”.

First of all, I know that you and your colleague spend a lot of time on the suggestions as you two are the main source of suggestions and comments. So I am sincerely sorry, that a lot of your suggestions also were part of the ones which moved to “not planned”. It was a conflict of interest, which had to be solved.
Regarding the renaming to idea portal: I think this way at least it doesn’t create an expectation, which can never be fulfilled. Creating a roadmap is a much more complex process than just executing top-down customer requests. One on the one hand side, e.g. neither Prometheus Exporter integration (34 votes), nor OpenTelemetry support (5 votes) are somewhere close to the top, but from discussions with many customers (including you) I know that these topics will become a major thing sooner or later. On the other hand side, the votes are in the end not reflecting the more than 4000 customers we have and are dominated by a few very active companies and also non-customers. Thus we have to make a qualitative judgement as well, when we include the suggestions from the idea portal. Thus, many sources need to be included when developing a roadmap for Checkmk.
Nevertheless, it helps us massively however, the more we know about what users wish for. That’s why the ideas portal is an important source for roadmap decisions.
More on this here: https://www.youtube.com/watch?v=VSy03_3tDXo

Some of the features with low votes which were implemented, were already planned anyway by us. It only made sense to also mark them as “implemented” then here.
Some features we plan, because we set ourselves a “theme” and then take in ideas which fit the theme (e.g. labels and working with folders in the 2.3 roadmap).

And this is among one of the reasons, why we need some kind of clean-up to make the entire feature portal also usable for users. At the current rate, we would have soon exceeded 1000 suggestions. And that was the clear wish from the community calls and discussions, that this is not something which can be handled by users, nor by us

The expectation of quality, especially from you, clashes with the concept of “hackathon”. The “hackathon” concept of the early days created a lot of technical and feature debt (remember service tags? A feature only half-heartedly designed), which we nowadays have to handle and reduce. Just adding another balcony on top of other balcony on top of 
 doesn’t provide a good product in the end. Also, in that case, proposed solutions would be implemented, while not thinking about the underlying problem. I will ask my colleague to share a good example on this with “fake check results should create hard states” and how a proper solution to the underlying problem can look like, once we have finished our usability tests.

Appreciate it. I would have also been happy, if you would have been part of the community calls :slight_smile:

2 Likes
19

Dear Martin

Thanks for your reply.

Part1

Just to clarify, because maybe I missed something?

  • E-Mail send out on the 11th of October announcing a meeting for the 17th of October
  • 17th of October for a Meeting

Thats it right?

Then I’m curious now about numbers and how to put this into perspective

  • how many paying customer particaipated in the call?
  • how many active users has the protal?
  • how many of your paying customers are using the protal?

Of course we would have participated but, just one option and running a time consuming 2.1->2.2 Major Upgrade, impossible. And I bet there’ve been others too.
Also at least for us, it did not feel that we have “voting power” on that meeting, rather then you just saying whats going on now.

Cheers
D.

20

I understand the need to clean up and I welcome it.

Your current “no votes in the last 6 months” rule leads to the absurd situation where good ideas with many votes are set to “not scheduled” just because most users voted for in the first days when it was released and other ideas with a smal number of votes remain in “Under consideration” because it got 1 vote in the past 6 month.

The status “not planned” implies that an idea is no longer being considered by you and will therefore not be implemented. So why should anyone still vote for such an idea?

Question
What is needed for such an idea to be considered again? 5 new votes since set to “not planned”?

Improvement 1:
As someone who has contributed a lot of ideas, I think it would be fairer and more comprehensible if all ideas with no votes in the last 6 months and less then 20 Votes are set to “not planned”.

Improvement 2:
There are ideas that I support because I think they make sense, but are not important to me at the moment, and other ideas that are very important to me, but get fewer votes from others because they only affect Enterprise Editions and therefore tend to get fewer votes from free edition users.

Perhaps this could be expressed with a kind of “golden vote” for Enterprise Customers. Each Enterprise Customers has three golden votes to indicate which ideas have the highest priority for them.

This would give you an additional weighting according to importance for the users.

1 Like