Future of Data Center Ask Me Anything (AMA)

128 comments

I have to agree with some members on this discussion about how disappointing this discussion has been. For something touted as being an all-day Ask Me Anything session, having the Atlassian members seem to be unavailable since 13:34 ET (10:34 PST) is very disappointing. his AMA session was announced on May 9th, 16 days ago and people responded with their questions right up until this morning. Including the first announcement of the discussion I count only approx. 16 replies from Atlassian. I know this seems a little petty to do a reply count, but I so wanted information to flow from Atlassian to us so I could make solid informed decisions about what my organization is doing and going to do with the various Atlassian DC products we have purchased.

I for one am looking to Atlassian to be a true partner with our organization. We see the value in the Data Center products and are working quite hard to expand on the offering to help our end users improve processes and productivity. It is extremely difficult to get their excitement up for a release, when all they see functionally are the Edit and Delete comment links moved from the top right corner to the bottom left corner of the comment field in Jira DC.

It would have been really helpful to have seen more information at Team 22 about the DC products, and including all of us as valued customers at the large annual user conference is a good way to ensure you have customers that rave about your products. Trying to fill in that gap via this approach was a decent attempt, but the execution is leaving a bad taste in our mouths (that's my observation,) and from what I see all it is doing is reinforcing the belief in people's minds that the Data Center products are not going to continue to have improved functionality.

I don't really know what to add into this. I was hoping for so much more, and believe that Atlassian is capable of much more when it comes to communicating to the DC customers. All of us purchased your products because we see the benefit to our organizations, and maybe Cloud is not the route we want to go or can go with, but having stagnant products is also not the route we want to go either.

Like • like this

Not to pile on, but to pile on ... :) I think Wayne articulated my opinion very well. I will just amplify one part. "We see the value in the Data Center products... All of us purchased your products because we see the benefit to our organizations."

In fact, it was only a few short years ago that Atlassian sold us on the promise of Data Center, which we fully embraced, merging 6 Jira Server instances to one Jira Data Center instance. This effort was expensive for us as a company, and we are now starting to reap the rewards. We are in no hurry to start over on a new platform.

So it is doubly disappointing that Atlassian has all but turned its back on its Data Center customers, essentially breaking an implied promise that Data Center is the preferred platform for large enterprise customers. It's not like it's been around long enough to be considered a "legacy" product.

Like • like this

I feel like I've partially wasted my time in taking time to articulate my thoughts here based on the pre-prepared nature of the responses. Maybe my expectations here were wrong but I was hoping for more of an open and honest dialogue around our questions.

Can this format please be raised internally for review? It's pretty tricky to track responses with no threading and I would prefer you don't do something like this again especially when the focus is data center but in many responses, the encourage is around moving to SaaS or talking about a SaaS feature.

Reviewing the comments, it would appear like you have angered your userbase through this process.

Like • like this

Hi @Lara Lopez . Thank you for sharing your feedback around the difficulty of managing custom fields. We've heard this feedback from other customers as well, and we know that the experience to manage custom fields at scale is not great.

We have also heard from customers that other aspects of administering Jira at scale need our attention. This is why we are planning to take a broad approach and analyse what dimensions in Jira could use improvement in terms of management. Based on this analysis we will make decisions on which problems should be resolved with the highest priority.

If you have any suggestions or feedback, we would love to connect with you to discuss further.

Like • Robert Pearce likes this

Hi @Matthew Martin,

Thank you for your feedback and we are aware of this limitation when it comes to the Archiving feature. To your point, the archiving feature was a good first step in helping reduce projects and issues on an instance and now we are exploring ways that we can improve archiving to further help with Performance and Scale.

I would also like to address the topic/theme of your question holistically, as we have stated one of our focus areas for future investment is around performance and scale. The concept of configuration falls squarely into this theme and therefore will be an area we review for improvement and optimization.

Like • John Dunkelberg likes this

Hi @John Dunkelberg , thanks for your feedback related to sustaining performance in terms of maintenance.
Performance and Scale is one of the main pillars of our focus and as part of that proactive measures to help make instances more maintainable is definitely part of the scope.

We have been working on Guardrails Documentation (coming soon). Guardrails are product based documentation and strategic recommendations designed to educate and help you avoid reaching a tipping point at which your instance might start to experience performance impacts. Additionally, we are also working on features such as Safeguards, coming with Jira 9, that proactively assist customers manage their scale.

Another example of planned work in this space is “Automated Cleanup for Projects and Issues” for 2023 (tentatively). When it comes to “View on Board”, it was redesigned to “Find on Board”, coming in Jira 9, fixing the performance issues.

Like • John Dunkelberg likes this

@Partha - Can you expand on what you are exploring?

Are there any areas you are specifically targeting and do you have any information about how you are looking to implement solutions? I've known that performance and scale have been a focus area for Data Center essentially since I've been using the tools.

Like • Robert Pearce likes this

Hi @Rachel Purpel , related to your feedback on documentation search. We’re aware of this problem, but there isn’t a quick fix as the functionality leverages a third-party search implementation that is failing. We are investigating whether we can replace it with something more robust in the near future.

Like • Dave Liao likes this

Hi @John Dunkelberg, While by default we allow a maximum of 12 users to edit a same page at the same time, an administrator can change this limit by using a system property. Please see this documentation for an overview of collaborative editing.

Please note that the usability of the collaborators badges in the top right of the editor may degrade with more concurrent editors, since there’s a limit to how many profile pictures we render in a single line. If you do make changes to the system property, we would suggest increasingly it incrementally, say to 15 rather than a significantly higher number.

Like • John Dunkelberg likes this

Hi @John Dunkelberg, thanks for your question. We definitely recognize the power of using Jira and Confluence together! Within our teams, we try to ensure that issues with integration points, such as the Jira Issue Macro, do get prioritised. It would be great to better understand where you are experiencing this friction between Jira and Confluence and if you have specific improvements that you’d like to see.

Like • John Dunkelberg likes this

Hi @kevin We’re aware that most of the products on the market have updated their Log4J lib to 2.17.x version and we will doing the same so continue watching https://jira.atlassian.com/browse/JRASERVER-62838 for updates on timelines (expect an update soon).

As you are probably aware, we maintain the private fork of Log4J 1.2.7 and make sure it is not vulnerable by patching it. Since log4j upgrade is a breaking change and not a simple task, thus we are planning its release carefully.

Like • like this

Hi @Gisela Lassahn, Automation for Jira is planned to become a native function in Jira Data Center in the near future.

Like • like this

HI all,

A number of you have mentioned your disappointment and for that I sincerely apologise.
We understand and are grateful that this comes from place of passion for Atlassian and we are lucky to have you as customers.

We strive to live our values and in the spirt of open company, I want to clear on Atlassian’s strategy.

Four or five years ago we looked at our position in the market, our customer base and our competitors and quickly realized that the future of our business would be won or lost in the cloud. Specifically in SAAS offerings. This was also the direction the majority of our customers are going.

For this reason we declared that we would sharpen our focus as a Cloud first company. We have followed this up by putting the majority of Atlassian R&D investment into developing our SaaS offering.

We understand that for some of our customers, particularly those in highly regulated industries, Data Center is still mission critical. That is why we are committed to continuing to support Data Center.

As the majority of our investment will go to cloud, we do not have plans to offer feature parity, between cloud and Data Center. As @Gosia Kowalska has stated, our future investment is focussed on Performance and Scale, Security and compliance and Infrastructure and operations as these are critical needs for highly regulated customers. This does not mean that features outside of these bounds will not be added, it is just that we will prioritise and assign the majority of our investment to these areas. Having said that we continue to evaluate all the customer feedback (like the many in this in AMA) and review our investment plans regularly. Thank you for all your feedback so far, and please keep it coming.

While I understand this is not the answer that many of you want to hear, it is important to us that you have clarity on where we will be investing.

Like • like this

Hi @Matthew Martin, specifically to next steps on archiving we are in early stages of exploring different options from data components that shall be available to be archived, to reducing friction for admins of when and how things are archived, potentially through some sort of automation.

We are introducing safeguards to help prevent performance and scale issues in the first place. For example, allowing admins to limit the number of comments on an issue and get notifications when issues are close to the limit.

If you are keen, I can connect you to the product manager who is exploring these topics.

I hate to say this, but if I pulled anything out of this AMA is this. Atlassian are stating they are now a Cloud first company and really have put Data Centre and all it's users on notice that it has a finite time left before it's put in the same place as server. I'm going to guess maybe 5 years left. And for those whom work in complex industries that require what Data Centre gives us (access to the back and, database, logs etc) and can't afford the double and triple the cost of Cloud will start the process now of looking for alternatives. So in short Atlassian seems to be willing to lose the Complex faithful customers that helped drive the business to cater to the non-complex user base so they can continue to be a point to contend with applications like Monday.com, Asana, Basecamp and the like. Sadly that means the take-away I see is start to have conversations with my Dev teams to see how invested they are in Atlassian and start working on an exit strategy for whatever the new application is. Most of here know our dev teams and have tested or continue to test the offerings of Atlassian Cloud and daily are disappointed on what Cloud offers currently. Wither it's the lack of support in general or even the lack expediting the fix (I spent over a week trying to recover a Confluence space that if was on-prem would have taken me two hours) to the Botched update that took down Cloud for users for 12 days, even knowing that devs want API and back end connections that it's not or was not thought about. If Atlassian really cared about it's Data Centre users, they would have used this AMA to listen to it's more dedicated users and actually put a project and plan together to work on solving what we need to continue; especially if this AMA was to build confidence that Data Centre isn't going anywhere. Instead, this AMA was Data Centre isn't going anywhere anytime soon, but we're going to make everything amazing in cloud and Data Centre will continue to suffer, until you decide to move to cloud. Which I can tell you my teams are already asking to move away from Atlassian and to something else.

Although I'm happy Atlassian decided this was hopeful of being a good idea. This AMA instead of building trust did the opposite. It proves that Atlassian doesn't care as long as they win in the Cloud.

Like • like this

@Jonny Carter (I hope I tagged the right Jonny)

RE: Java 17.

We are planning the rollout of Java 17 across Data Center. We will balance this with the deprecation and removal of Java 8 support. Some key considerations in this for us are the cost and timelines of the underlying Java support for our customers.

Jira: Java 17 is planned and we are working on it on the 9.x branch. Dropping support for Java 8 will be considered in a similar timeframe. This also applies to Jira Service Management 5.x.

Confluence: Confluence 8.0.x drops support for Java 8, keeps support for Java 11 and adds Java 17.

Bitbucket DC: We are looking at support for Java 17 as we speak and need to determine which release to drop it into.

I am talking to the architecture team about your Servlet question and will get back to you ASAP!

Like • like this

"our future investment is focussed on Performance and Scale, Security and compliance and Infrastructure and operations as these are critical needs for highly regulated customers."

@Partha @Gosia Kowalska @Niall O'Riordan Again here is this saying. And again, I'm asking where the market research says this is all people want. You keep saying this like it's some mantra that will ward off questions about the roadmap, but I still can't see anyone asking for this to be prioritized over anything else.

We've had performance updates for years now. It seems like there is some statement about improved performance in every release note since Jira 8 came out. Scale has been there too. But what about accessibility? Dark-mode isn't just a nice-to-have for people staring at monitors all day. This high-contrast mode can be critical for low-vision users. But the message I'm hearing is that because they are on Data Center, they don't count.

That's not even mentioning the other features that can make people's lives 10000x easier. I have been a champion in the community for Jira Work Management since before it was officially released. This seems like a no-brainer for companies large enough to need Data Center products. And let's face it, with the current hard limits on Jira Cloud - your largest deployments have to be on Data Center. Those are the users who you are letting down. And they don't complain to you; they complain to us, the Jira Admins in their orgs.

My original post on 4/13 on https://thejiraguy.com , I stated that your quietness around Data Center was a sign Atlassian was no longer interested in supporting it. I got you talking about Data Center, as your response and this AMA prove. Even noted so in a follow-up. But your roadmap - OK, let's be honest, you're putting Data Center on essentially what amounts to a Maintenance Mode - still supports my original assertion. I have seen nothing in this AMA that convinces me that Atlassian is a viable long-term partner for Data Center. Every question about new features or the roadmap is answered with this same response. And I don't know who in marketing told you to chant this, but they are doing you no favors.

Like • like this

Well said @Rodney Nissen - ReleaseTEAM , i think it’s to for DC users to plan their exit strategies, planning staying a DC customer long term is looking more and more to be a big mistake.

Like • like this

Absolutely @Rodney Nissen - ReleaseTEAM

As a Jira Admin, I receive a lot of questions about featurss on Jira and I end up saying things like: "It's coming soon", "I'm not sure we can do that yet", "I'll have to check with Atlassian".

That should not be the norm! I should have the confidence knowing features requested are taken seriously. This AMA is an eye opener for sure but you only have to look at the feature request backlog to really know whats going on.

Some requests have been YEARS in the making. I've seen a few 5, 10 years old. Really small, subtle features that make our lives easier.

It's not easy being a Jira Admin but some of these simple features would go a hell of a long way!

Like • like this

I want to thank Atlassian for conducting this AMA. Others are clearly disappointed and a few seem to think Atlassian's statements about the future of Data Center are dishonest. I actually feel the opposite. The statements from Atlassian lately (and repeated here) are consistent with the facts on the ground and the features delivered for the past couple of years. If anything, Atlassian is being more honest now than they were when they first decided to go all-in on the cloud (which was more than 4-5 years ago and well before any public announcement).

I don't blame Atlassian for going cloud-first. It's a business strategy, and clearly one that's working for them as a company. I think the disconnect is really one of perception. Atlassian continues to make aspirational statements about Data Center, clearly believing they can be cloud-first and also do right by their Data Center customers. For a number of customers (myself included), their delivery track record and statements about future investment indicate the Data Center products are in maintenance mode.

At the end of the day, Atlassian is going to do what makes sense for their company and we're going to do what makes sense for ours. The Data Center products are good products and for now, it makes sense for us to use them. Based on how things are headed, we'll probably reach a tipping point down the road where it will make sense to move. Nothing here has changed that, so the question is really when it will happen and what we'll move to. I think Atlassian is hoping their cloud offering will be good candidate when we reach that point.

Like • like this

Thanks @Partha - we've had stability / performance issues with 12 collaborative editors, so what I'm interested in is Confluence increasing its scaling capability, not how to up that limit in the existing system (though that's interesting to know). Hopefully this will fit in your Performance and Scale theme.

I know you have a lot of comments here, so I'll reproduce what I said in the original comment for context:

"We've had great adoption of Confluence during the pandemic, including for some high profile senior management meetings. However, we consistently hit pains with collaborative editing in terms of stability and the current 12 simultaneous editor limit. As much as I love and advocate for Confluence usage, particularly because of its interconnection to Jira, I will have to encourage investigation of alternatives such as Microsoft solutions if Confluence cannot operate confidently at a higher collaborative scale."

p.s. I hope you folks can work on getting the name-tag functionality smarter here. On our Confluence instance it seems to be clever to suggest the right people, but on this support instance it's hard to tag the right person, particularly with common names.

Like • Dave Liao likes this

Looking forward to your summary @Atlassian

With regard to the fact that documentation search on the support site for Data Center does not work *at all* and there appears to be no urgency to get it fixed, might I suggest that you remove the search box altogether?

https://community.atlassian.com/t5/Data-Center-articles/Future-of-Data-Center-Ask-Me-Anything-AMA/ba-p/2023869/page/5#M445

https://confluence.atlassian.com/jirasoftwareserver0820/jira-software-data-center-and-server-8-20-documentation-1095248231.html