--- layout: markdown_page title: "Atlassian Bitbucket" --- ## On this page {:.no_toc} - TOC {:toc} ## Overview
<%= partial "includes/atlassian_comparison_tabs_vue" %>
## Summary Atlassian Bitbucket gives teams Git code management, but also one place to plan projects, collaborate on code, test and deploy. It is marketed in the SaaS form (Bitbucket Cloud) and in a self-managed version (Bitbucket Server), however they are not the same product. Bitbucket Server is simply a [re-branding of Stash](https://www.atlassian.com/blog/archives/atlassian-stash-enterprise-git-repository-management). The two products are completely different code bases, written in two different languages ([Cloud in Python, Server in Java](https://en.wikipedia.org/wiki/Bitbucket)) and do not maintain feature parity. Because of separate codebases they each have a completely different API, making it much harder to integrate. Bitbucket supports Mercurial or Git, but not SVN. GitLab does not support Mercurial or SVN. GitLab is a complete DevOps platform, delivered as a single application, with built-in project management, source code management, CI/CD, monitoring and more. Bitbucket only does source code management. You would need to use Atlassian Jira to get project management, Bamboo for CI/CD and Atlassian does not provide a monitoring solution. Additionally, GitLab Ultimate comes with robust built-in security capabilities such as SAST, DAST, Container Scanning, Dependency Scanning, License Compliance, secret detection and more. Bitbucket does not support these capabilities and Atlassian does not have a product for them. GitLab also offers an "on-prem" self-managed and "cloud" SaaS solution. GitLab runs the exact same code on its SaaS platform that it offers to its self-hosted customers. This means customers can migrate from self-hosted to SaaS and back relatively easily and each solution maintains feature parity. GitLab has seen an increased interest from [world-wide internet searches](https://trends.google.com/trends/explore?date=today%205-y&q=bitbucket,gitlab) (a strong indicator for devops interest) over the past five years. Between 2018 and 2019 Gitlab adoption as a Version Control System has increased by 21% whereas Atlassian Bitbucket's adoption has decreased by 11%. This is as per [The Next Stack's analysis](https://pbs.twimg.com/media/EGoXqXeXUAE_Dva.png) of Jetbrains Developer Ecosystem surveys. ## Gaps * Extending the native functionality of Bitbucket is done through plugins. Plugins are expensive to maintain, secure, and upgrade. In contrast, GitLab is [open core](https://about.gitlab.com/2016/07/20/gitlab-is-open-core-github-is-closed-source/) and anyone can contribute changes directly to the codebase, which once merged would be automatically tested and maintained with every change. ## Response to Atlassian's Bitbucket Comparison Page Atlassian has a [Bitbucket vs. GitLab comparison page](https://bitbucket.org/product/comparison/bitbucket-vs-gitlab). This update is to ensure that the correct product offerings are being compared (i.e., GitLab.com, GitLab Self-hosted, Bitbucket Cloud, Bitbucket server, etc.) * Purpose built for the cloud - the site claims that GitLab is not purpose built for the cloud, which is somewhat misleading. GitLab is a single app that was built to run on many different environments including "the cloud" with GitLab.com. Additionally, GitLab actually does zero downtime deployments. Also, Bitbucket Cloud and Bitbucket Server are different architectures (i.e., code bases) and diverging in functionality. See [Bitbucket Rebrand FAQ (updated Jan 2018)](https://confluence.atlassian.com/bitbucketserver/bitbucket-rebrand-faq-779298912.html?_ga=2.68100121.980444013.1542263978-1592705151.1523946360) for more information. * Best in class JIRA integration - It makes sense that Bitbucket should have great JIRA integration. GitLab is an integrated application from planning to monitoring with GitLab issues being a powerful tool for idea management, iteration planning, issue tracking and software development. However, we recognize that many organizations have been using JIRA for many years and have existing data and business processes built into it. Our goal is to make GitLab work seamlessly with JIRA, offer the best integration possible, better than the current Atlassian suite provides. Check out our [JIRA integration page](https://about.gitlab.com/solutions/jira/) for more details. * [Code Aware Search Beta](https://bitbucket.org/blog/introducing-code-aware-search-for-bitbucket-cloud?_ga=2.131692365.35318360.1558109698-1703890985.1558109698) - code aware search is a feature that ensures definitions matching your search term are prioritized over usages and variable names. Although GitLab does not currently have this feature, we've made code search even better with the [GitLab 11.1 release](https://about.gitlab.com/blog/2018/07/22/gitlab-11-1-released/). With enhancements to the advanced search syntax, you can now filter by filename, path, and extension, enabling you to find the right files quickly. * Mercurial Support - True. See the Summary section above. * Bitbucket UI add-on integrations - N/A ## Comments/Anecdotes * Evidence that Bitbucket Cloud and Bitbucket Server/Data Center are two different products and Atlassian is focused on the cloud one: * From Atlassian themselves, Bitbucket Cloud and Bitbucket Server are different architectures (ie. code bases) and diverging in functionality. This is more or less spelled out in their [Bitbucket Rebrand FAQ (updated Jan 2018)](https://confluence.atlassian.com/bitbucketserver/bitbucket-rebrand-faq-779298912.html?_ga=2.68100121.980444013.1542263978-1592705151.1523946360) * From a recent [HackerNews article](https://news.ycombinator.com/item?id=17909454) >Atlassian wants you to move to their cloud product, it's that simple. Server is a thing that makes their life harder and stops them from extracting maximum revenues from you so you can expect a soft but persistent push to switch to cloud which will get harder over time. * Disparity between Bitbucket Server and Bitbucket Cloud for a top customer requested feature took Atlassian over 2 years to acknowledge. Customers found out the feature existed in Server but not Cloud once moving to Cloud. [Still not resolved as of Q4 2018](https://bitbucket.org/site/master/issues/12833/branching-models-for-bb-cloud#comment-45982415) * [An analysis of Bitbucket Server](https://docs.google.com/spreadsheets/d/1BnxRkNKytRcEnKqlnCyJFy_B6swjGyLdcwGi9bg32PY/edit#gid=0) new features per release and feature-to-maintenance ratio, per release since it was changed from Stash Server, shows a downward trend in innovation ([Data derived from Atlassian Bitbucket Server Release Notes](https://confluence.atlassian.com/alldoc/bitbucket-server-documentation-directory-278071997.html))
Bitbucket Server New Features Graph
* Discussion from [HackerNews article about Atlassian not allowing benchmarking](https://news.ycombinator.com/item?id=18103162#18103813) > Atlassian has always forbidden to talk about the performance of their products in their ToS and in their previous EULA. We all know why, but we don’t talk about it. ## Resources * [Atlassian Bitbucket Website](https://bitbucket.org/product) ## Pricing - [Bitbucket Cloud](https://bitbucket.org/product/pricing?tab=#tab-5834cfd9-fbc0-4f21-ae10-65832598b12a) * Free tier - $0 - Unlimited private repos, Jira Software integration, Projects Pipelines (50 build mins/month), 1GB/month limit on file storage * Standard tier - $2/user/month (min $10/month) - Same as Free + 500 build mins/month + 5GB file storage/month * Premium tier - $5/usr/month (min $25/month) - Standard + some advanced features + 1000 build mins/month + 10GB file storage/month - [Bitbucket Server / Data Center](https://bitbucket.org/product/pricing?tab=#tab-db2f88e8-2388-4417-8e4f-6a7205437cc7) * Server - starting $2k perpetual (25 users, ppu drops roughly every 2x previous tier), includes year maintenance, single server, unlimited priv+pub repos * Data Center - $1800/yr (25 users , ppu drops roughly every 2x previous tier) includes annual maintenance, Server + HA, DR, mirroring, SAML 2.0 * Must buy Data Center if over 2k users. ## Comparison