That's a wrap: AMA with Taylor Pechacek on new developer tool integrations in Jira Software Cloud

[Taylor]

Update: That's a wrap! Thank you so much for your thoughtful questions. I will keep an eye on this post and try to answer any additional comments in a timely manner. 

Hello Atlassian Community,

My name is Taylor Pechacek and I’m Atlassian's Principal Product Manager for Jira Software Cloud.

On Thursday, May 16th from 3-5 pm Pacific, I'll be hosting real-time AMA to answer your questions about developer tool integrations. Here's how it works:

Add your questions below any time before or during the AMA. Be sure to take a look at other community member’s questions and up-vote those that you find interesting. Amongst other topics, you can ask me things like:

• What are the best practices working with this information
• What's next in developer tool integrations
• What other partners are you working with

At 3 pm PST on Thursday you can expect to see answers from me and my team rolling in. Watch the page and be ready to add follow-up questions and discuss further with other Community members. We'll wrap the event at 5 pm PST, but will be sure to answer any questions we didn't get to.

Cheers,

Taylor

[Jack Brickey]

Similar to @Andy - PTC Redundant we don't currently use any development integrations. Our development is limited to a handful of internal tools, improvements to our ERP solution, an occassional EDI with customer's software asset systems and some web portals. What integrations might benefit us and why? Secondarily, what hurdles might we face as we begin to deploy these integrations?

[Taylor]

Hey @Jack Brickey, check out the answers above as well.

As for a great place to start. I always recommend starting at connecting your code with the work being tracked in Jira Software Cloud. Having this extra visibility about what it means to be 'in progress' helps give so much more context to everyone on the team.

From the opposite side, let's say you are doing a code review. Having the related Jira issues in Bitbucket as well helps you understand why and what the code is trying to get done. That context help speed up code reviews and team members give much better feedback in the code review itself.

Also, if you are deploying that code often, then you could start using Bitbucket Pipelines. This works automatically with no setup if you already have Bitbucket Cloud and Jira Software Cloud integrated. But you definitely don't need to start with this as well. Here is some more context on that if you are interested: https://bitbucket.org/blog/jira-software-bitbucket-pipelines-new-integration-gives-continuous-visibility-from-backlog-to-release. 

Lastly, on the hurdles you might face. Honestly, the biggest one is just getting the team to adopt the behaviour to add issue keys to your commit message & branches. So, 'TEST-123 Finished the unit tests". There are tools that can help do this for you automatically in your IDE. https://bitbucket.org/blog/how-atlassian-for-vscode-changed-the-way-we-ship-code is a great example of this! 

However, once you do this, the rest will automatically be synced to Jira. If you add issue keys to commit messages then all the PRs will be synced as well that include those commits. And that's how Bitbucket pipelines knows how to automatically send related builds and deployments if you set that up as well.

(Bonus) The new Jira Software Cloud next-gen projects have a really cool feature that let you see the development status right on the cards! This gives you a really clean, simple way to view what's happening at the Board level across the work that is in progress. 

Let me know if you have any other questions, I would love to help you get started here!

[Andy - PTC Redundant]

Yes! I can relate. We're totally not tracking. 
We had an intern last year that could do exactly all of this but then he left and despite a Knowledge Sharing session and a well documented Confluence wiki, no one cares enough to pick it up....! Even my own enthusiasm has wilted. :'o(

[Tinker Fadoua]

Hi @Taylor 

In Summit, there was lot of exciting information regarding AWS and JIRA Cloud. Do you have a roadmap on what is already available and additional changes happening soon?

I am very interested specifically about AWS and JSD.

Best,

Fadoua

[Taylor]

Hey @Tinker Fadoua, thanks for the question!

For Summit, we did actually launch a workshop guide which shows you how to use a few different AWS services and the new Jira Software Cloud dev info, build & deployment APIs. However, it's purpose was more of a prototype and guide to show what is possible, check out https://atlassian.awsworkshop.io/ for the details.

Specifically, this means that the workshop assumes you would set up your own 'app', using the AWS infrastructure to host it. Once it was set up and wired up to both AWS and Jira, it would start sending data and you could see deployments, as one piece of data, right inside the issue detail view. 

However, this means there is no single integration that you could go easily download and install from the Atlassian marketplace. We are working with them to find the best path forward to productionise the workshop essentially, but no definitive timeline on that yet. This would handle the integration between many AWS customers and many Jira Cloud customers.

Lastly, re: the announcements with Jira Service Desk and AWS, I would need to sync up with a few colleagues in JSD to see where that is at, since it is a bit different than the Git and CI/CD focus we have in Jira Software at the moment.

Cheers!

[Luis Schweigard]

Hey @Taylor !

I'm loving your plugin API for Confluence, i just have one issue using ActiveObjects. When performing an operation in the ActiveObject.executeInTransaction()-method everything works fine, but once I try it with the @Transactional-Annotation, the ActiveObject is created but not persisted.

Any idea how to solve that issue? The only dependencies I've added are 

com.atlassian.activeobjects

and 

com.google.collections

Thanks in advance, you could really make my day / week here ^^

[Fabian A_ Lopez _Community Leader - Argentina_ Florida_ California_]

One of the concerns we have by selecting Cloud instead of Server is security, specially HIPAA compliance. On the previous AMA about security was mentioned that HIPAA is part of the cloud strategy. Are the integrations to development tools considering that aspect as well?

[Taylor]

Hi @Fabian A_ Lopez _Community Leader - Argentina_ Florida_ California_ - you are correct, HIPAA compliance is on the roadmap for our Cloud products. You can view the complete compliance roadmap here.

The developer tool integrations that we have launched focus on the software delivery pipeline, e.g. building and deploying software applications. These Integrations are not transmitting or storing PHI so are not exactly within the scope of HIPAA compliance. 

That being said, we do take security and access controls very seriously for any application that integrates with Jira Cloud. Any integration that we launch on our marketplace goes through a security review and only authenticated admins can install a new integration.

Lastly, the permission model for these integrations is tied closely to the Jira Cloud issue permissions. So, if a user does not have access to see a specific Jira issue in a project, then they will not have access to the related code, build, deployment, or feature flag inforamtion. Therefore, any improvements we get for security and compliance across Jira Cloud these integrations would get the benefits as well!

Let me know if you have any other specific questions.

[Andy - PTC Redundant]

@Taylor : Our company doesn't do any integrations at this time.
We use cloud "lightly" (30 users, not intensive Jira users) and it annoys me... I want to PUMP UP the use rate and get integrations working to our advantage!
What are the top 3 (or more) reasons that a company like ours (highly innovative, ultra agile and lean) needs to take their integrations more seriously!?

[Taylor]

Hey @Andy - PTC Redundant, amazing question!

So for a bit more context, how would you describe your 'stack'? I'm assuming you have a tool for managing source code, probably some CI/CD tools, and maybe a feature flagging service?

Top 3 reasons

1.  If you are a small, ultra agile & lean organisation, then you are probably quite comfortable having information be open and visible to everyone on the team. This gives everyone the context & autonomy to make decisions quickly. Starting with the Development Information integration, anyone (who has access to that issue) can see commits, branches, and pull requests on issues in Jira. The visibility here is really powerful and helps facilitate much better conversations. Lastly, anyone can click on that pull request and be taken directly to that pull request into the source code tool you are using. Having these connected links help you navigate across products really easily. 
2. When using integrations to wire up commits, pull requests, builds, and deployments, it can alter how you run standups and keep the team focused on key problems & removing bottlenecks. We see a ton of teams like yours set up the Dev Info integration with Bitbucket or GiHub and add a quick filter on their sprint board "Show me all the open Pull Requests" (You can use JQL with this data!). They focus in on those areas (maybe they need code reviews, or forgot about merging it, etc.) during the standup, while keeping the context about the what/why of the problem.
3. Lastly, the deployment and feature flag integrations with other Cloud SaaS providers are quite new. We had Bamboo + Jira Cloud previously, but this has expanded to many different providers and will continue to grow. The benefit here is that you can quickly see which features were deployed to customers, or a staging environment you are using to test. Being able to create an epic (representing a feature perhaps), then breaking that down into all the individual stories is a nice way to work on that feature across several developers. But what makes this even better, is now you know which ones are actually deployed to production. So you know when it's ready to test, or release to your customers. 
4. (Bonus) The integrations are quite easy to set up and get started. You may have to change a bit of behaviour and your process of development but it can bring tremendous value down the road. Specifically, I mean you can use this information to start automating your workflow and stop having to manually keep Jira up to date. You are a smaller team and want to move fast, so having this additional context about dev, builds, deployment, feature flag info can really start to speed things up. A popular one is "When a pull request is merged, move it into a specific status (maybe it's 'ready for release' for your team - up to you on what you want to call it).

Let me know what you think! What are the top 3 problems your team faces at the moment (regardless of whether it is related to integrations or not)?

[Taylor]

[Andy - PTC Redundant]

Thanks so much for the awesome answer @Taylor 
Wish we could have done this more live in during the AMA but i'm not on PST :o(
The question you posed probably sums up the entire challenge I face: my answer is I just don't know ! :(

I am going to take this 4 points to heart and work on them, plus delve into more detail with the teams so as to be able to answer with more than just "I don't know!". Will come back on that next week.

I would probably say that at a very deep level we face the current challenges (of which the issue with integrations stems from):

1) Lack of collaboration.
2) Silo behaviour (yes! Even in such a small company. Craaaazy!).
3) Absence of openness, visibility in each other's work and knowledge sharing.