This sounds like a terrible move and will probably force all medium-size organizations I work with to ditch Atlassian.
We have the technical know-how to administer data center versions of their products, but we can’t do shit if they force potential customers to pay for a minimum of 40 000 USD for a license per product.
Many of these orgs have multiple Atlassian products so this will probably end up doubling costs on several products at the same time effectively making it a money sink and impossible to justify to their budgeting.
I would have been perfectly OK with DC taking over server with similar pricing, but this is just a monumentally idiotic greed-driven move.
Now small-medium-large organizations are either forced to pay A LOT of money for self-hosting their highly protected data or either use their cloud. The latter is out of question for almost all of the organisations I know.
If I was a Microsoft or another enterprise tech company I’d hire a thousand engineers tomorrow to develop a Jira/Confluence competitor before the grace period for server licenses ends. All that engineering money will pay itself back whent they can sell modern collab tools without tech debt from 15 years ago for a price we can tolerate.
Wow. This must feel like a huge fuck-you to people that migrated to Atlassian because they could host things locally (I worked at a place that made the leap 3 years ago and it was a somewhat controversial decision. I feel kinda bad for everyone that'll be hearing an "I told you so" on Monday morning).
Not only is the product being killed, they're raising the price during the migration period. Their FAQ also hints that even more price increases are coming. I imagine they'll lose a lot of customers that can't or won't send all of their data into the cloud
This is entirely because they had to fork their product and codebase between cloud and on-prem, and their resulting total failure on execution on their internal product roadmap.
Atlasssian has demonstrated a chronic inability over the past few years to maintain feature parity between what are effectively two different products.
Atlasssian has had a really, really terrible feature execution delivery pace over the past few years - "Next-Gen Projects" were so immature at launch they were useless, and have been incredibly slow to mature.
So they need to focus their developers onto building features on the Cloud product codebase, but I am not hopeful that this means they'll actually ship features faster.
It's insane that they had to that rather than giving their On-Prem version the ability to scale the way you need a SaaS service to scale. Turning your SaaS Ops team into an On-Prem customer is a great way to dog-food your product.
We're using Jira, Confluence, and BitBucket on-prem at our DC now, on some pretty sizeable VM's with good network connectivity back to head office.
Despite that, it feels very slow. Not the occasional 10+ second page load, where you've accidentally clicked something that happens to generate a cascade of cache misses. I can forgive that kind of infrequent performance hiccup.
But rather, regular usage, especially of Confluence and Jira, just feels sluggish when doing bread and butter actions - adding a comment to a ticket or reloading a kanban view, loading a new page or opening the search dialog on Confluence.
Yet, everything I've heard about the responsiveness of the 'cloud' offering is that it drives people to move to the on-prem server version.
Atlassian does not provide particularly good guidance on how to get their tools to perform well. The first big issue for any moderately busy system is that you need to dedicate huge gobs of heap memory (think 16G+) to the tool. After that, they suffer from poor schema design - there are tables that are literally missing indexes on commonly queried columns, and some columns that have delimited values embedded in them.
It's hard for me to imagine that atlassian would allow individual instances of jira/confluence/etc in their cloud versions to get nearly as much resources as you need to give them on-prem. The incentive would always be for them to try to minimize the amount of memory every instance gets, which does NOT maximize your individual performance. Shared resources will also make performance more variable and "noisy", making it more difficult to narrow down specific causes. It's hard for me to see how "Java" and "cloud" go together.
I've no recent experience of the on-prem version, but nowadays I have to use the cloud version and it's painfully slow. One common problem I encounter is that after a page loads I might want to search for something in the search box. I'll click it and start typing ... then at some point the rest of the page has loaded including the stupid key bindings which do various random actions on different keystrokes. I don't know why they have this, it's infuriating.
Oh and they keep updating the UI which involves shuffling things around and every new version is slower than the last. If you're running on-prem you at least get to decide when to update and notify everyone accordingly. We'd just find out one day that everything has been moved around and learn the new locations of different features.
Sorry I'm really just generally ranting about Jira now but honestly I find it so infuriating to use. Confluence as well, how did they manage to make wikis suck so bad?
Yep:) At a previous job, we also did Jira/Confluence/BitBucket self-hosted, running on what should have been nice hardware, and spent months chasing performance issues, including a charming issue that was so bad that it broke jira for the entire company every... Monday, I think? When you have a well-used standard operating procedure for restarting Jira just to unfreeze it, you have a problem. We were pretty sure it was somebody running a weekly report with some monstrous query, but 1. not even Atlassian support could seem to tell us how to find such an event, and 2. I don't really care how bad the query is - if you can run a query and take down the system for the whole company, then the software is bad. And honestly, even "normal" performance was poor - I seem to recall that it at least acted like the web UI was the slow part; running a local CLI client against the API was faster.
The performance on the cloud version is awful. When I previously self hosted it was a miserable management experience, but it definitely felt snappier than the current cloud version does.
Jira is probably the slowest piece of software I use in my job. I can't believe there's an interstitial loading screen with a status bar for cloning an issue.
I don’t share the same experiences with slowness (maybe we do something right on the instances we manage). I do however experience slowness in cloud. I assume they increase cloud resources in some kinds of steps that make them appear extremely sluggish.
While we had very similar experiences - only used JIRA but the performance was borderline-unusable - I've heard reputable sources say that configuration plays a tremendous role in that. I'm inclined to believe we just hadn't configured it "correctly," but the fact that you can destroy performance by casual misconfiguration is itself a pretty big issue.
I've heard the same about cloud performance, which seems to be an argument against it being simply configuration, but who really knows?
Ouch, on-prem software is a really tough business. I can kinda see why companies want out. Especially as they get bigger. Two of my former employers had to go through this hurdle of moving to SaaS from a an on-prem installation.
* It's really hard to scale support. You end up being on the hook for why "it's not working" in thousands of different of environments you have zero control over. Anyone qualified to do that level of support could be making more money not hating their life.
* Nobody wants to pay for updates. I mean it's totally rational but we also don't want to support $old_version forever.
* Even when updates were free people still didn't do it. Telling your users "sorry you're using an ancient version, please update" just makes people angry. Again, understandable, but surprise! those same people suddenly weren't angry about updates once they were on our SaaS offering. Turns out that frog boiling is reasonably effective.
* We had to deal with swaths of customer complaints about performance that we're mostly not our fault. Sorry our app is slow on 1/64th of a used SuperMicro. Maybe your IT department should get on GoFundMe?
This title is misleading. Atlassian will continue to sell and support their Data Center lineup, which are all on-premise products, so they are not moving to “cloud-only” as this title suggests. Now, this announcement will probably affect SMBs the most, as most large organizations, government agencies, etc are most likely already running Data Center and have left Server behind since it really isn’t scalable. I would’ve assumed that they would have just jacked up the price on Server to squeeze everyone out, but alas here we are.
It's an difficult position Atlassian are inflicting on tens of thousands of customers. Atlassian's self-hosted products are uniquely flexible, being built on a plugin architecture, and many orgs have indeed customized Jira extensively with plugins, notably ScriptRunner [1]. Atlassian's Cloud plugins have their APIs, but have nothing like the same flexibility. A lot of functionality just isn't possible in the Cloud architecture. It's a bit like Firefox moving from XUL to an extension API.
In my opinion, the most customer-respecting way forward would be for Atlassian to open-source their discontinued Server product line. Go to the "open core" model with the clustered Data Center product as the upsell. This avoids screwing over their customers, and if their Cloud product really is as good as they say, customers will migrate to it naturally over time. It's the kind of damn-the-torpedoes move I think Mike CB would like.
But over the next 3 years, lots of painful migrating or evaluating-of-alternatives will need to happen, and perhaps a non-Atlassian forum for sharing experiences will help.
I promoted it upthread (though I have no connection to JetBrains), but I'd note that YouTrack and their other team products have extensive Java and REST plugin APIs so they're also very customisable. Actually YouTrack lets you write workflow logic in JavaScript too.
> open-source their discontinued Server product line
little did you know, the source code for atlassian server products is already available for any paying customers. It's not FOSS, but if you need to run this yourself, and customize it, it's already possible. But of course, you will be responsible for keeping it updated (patching vulnerabilities from libraries etc).
Plugins are great until you crash your standup board because the “Recycle” plugin put stories somewhere you can’t see but also didn’t remove the story from the sprint, causing permission issues.
A lot of what makes Jira so broken is that it is way too flexible, and this flexibility has costs in terms of performance and stability. There’s a damn good reason why I used Chrome back in the XUL days for FF; because XUL FF was not performant or stable for me back then.
My employer (large Fortune 500) recently "soft banned" Confluence and Jira, meaning groups already utilizing it can stay on it until our enterprise license runs out at the end of the yearly support contract. At first I thought it was really stupid decision as Confluence, at least, is fantastic product and Sharepoint/Notes (the enterprise replacement solution) is a horrid alternative. One of the admins said there were licensing issues that drove the decision. I figured someone was just being cheap but I bet this was the reason.
Data sovereignty means my 5,000 seat company will stop using it. Confluence and Jira were already a hard sell due to licence fees. Now Atlassian wants us to put our trade secrets on AWS or something? Not happening.
At least they if we mess up security on a host inside our business it’s not the end of the world.
edit: I don’t make the money decisions here, I am just a peon.
Server products is what made us purchase Atlassian software. There are numerous reasons why our company (and a lot of other companies I presume) would like to avoid cloud. And if we really had to go cloud, I simply don't see why we would choose Atlassian over alternatives.
This is a big problem for our non-profit organization. All of our documentation is stored in Confluence. We are fortunate enough to use a free Community license for self hosted servers, but this announcement would remove the free Community license and would force us into a Community Cloud license that would cost us over $4k a year. Are there any good, possibly open source, alternatives available for non-profits?
How does this play with the Australian ruling around state mandated backdoors and encryption bans? I can't imagine there's a lot of trust here for an Australian company providing cloud infrastructure that can be compelled to include back doors or locking out encryption?
I work in SaaS supporting global regulated markets. We’ve had to replicate full stack implementations hosted in local data centers and accessible only by citizens physically in that country with security clearance. I expect Atlassian (and other cloud companies) will need to do the same.
I’ve abandoned Jira (cloud) at work so long as my manager continues to not care. It’s just painful. Brutally painful. Like 20MB download per use and everything takes forever to populate and there’s lag in everything I do.
Atlassian feels like a bunch of individually purchased products that they then just jam together with integrations that never quite work the way they should.
I would guess that this move is because because their software runs like a bag of hammers falling down a flight of stairs and the last thing they want to deal with are your support calls about why their products dont work right. Don't get me wrong, they can work smoothly- just it takes an order of magnitude more work than you would expect / they would admit.
To be slightly more charitable, this is a sensible move as at least this way Atlassian can control the end user experience and (perhaps!) have it not be utter trash. Far easier for them to manage this way.
I maintain server versions most of the self-hosted applications they're killing off.
One thing I like about the server versions is the ability to spin up development versions of these applications so I can test out new plugins, version updates, or complicated configuration changes without affecting our production applications.
So much for that. I guess I can skip that step and when it goes wrong in production I can be the old man who yells at "the cloud".
I'm using it for ages now and i have not seen any real features being added in the last 10 years and fundamental issues have not been adressed.
The weirdest thing was their UI change a few years back where they thought it would be great to make core workflows 1 or 2 clicks further away than before.
Like 'creating a ticket; went behind this plus button thing.
Plenty of features i would love to see, the community votes them up as well but nothing happens.
Apparently they are to busy with stuff no one is seeing. Hope that actually brings in innovation.
Everyone keeps only mentioning gitlab, whereas gitea[^1] has wiki, bug tracking, and git as well. No CI if I understand it correctly, but for the rest, it's there, and it's nice, and it's order of magnitudes faster and lighter.
Then there's fossil[^2], which is brilliant, and is completely self-contained, with version controlling, wiki, etc. It's a bit awkward, but nonetheless feature full.
Switched from Jenkins to Bamboo Server about 5 years ago. The integrations with Bitbucket, Jira, and Slack have been really useful. My team has put a lot of time into getting the most out of Bamboo, so it’s disappointing that we won’t be able to continue to build on that.
It would be nice if Atlassian open-sourced Bamboo so that we could continue using it long-term, but honestly one of the primary reasons that we’ve stuck with them is their excellent support team. Without them, our development team would be forced to diagnose and fix issues in Bamboo, which is not the best use of dev time. So I guess we’ll attempt to find alternative build software soon.
[+] [-] originalvichy|5 years ago|reply
We have the technical know-how to administer data center versions of their products, but we can’t do shit if they force potential customers to pay for a minimum of 40 000 USD for a license per product.
Many of these orgs have multiple Atlassian products so this will probably end up doubling costs on several products at the same time effectively making it a money sink and impossible to justify to their budgeting.
I would have been perfectly OK with DC taking over server with similar pricing, but this is just a monumentally idiotic greed-driven move.
Now small-medium-large organizations are either forced to pay A LOT of money for self-hosting their highly protected data or either use their cloud. The latter is out of question for almost all of the organisations I know.
If I was a Microsoft or another enterprise tech company I’d hire a thousand engineers tomorrow to develop a Jira/Confluence competitor before the grace period for server licenses ends. All that engineering money will pay itself back whent they can sell modern collab tools without tech debt from 15 years ago for a price we can tolerate.
[+] [-] Rebelgecko|5 years ago|reply
Not only is the product being killed, they're raising the price during the migration period. Their FAQ also hints that even more price increases are coming. I imagine they'll lose a lot of customers that can't or won't send all of their data into the cloud
[+] [-] Martin_Beck|5 years ago|reply
So they need to focus their developers onto building features on the Cloud product codebase, but I am not hopeful that this means they'll actually ship features faster.
[+] [-] ashtonkem|5 years ago|reply
[+] [-] x3n0ph3n3|5 years ago|reply
[+] [-] Jedd|5 years ago|reply
Despite that, it feels very slow. Not the occasional 10+ second page load, where you've accidentally clicked something that happens to generate a cascade of cache misses. I can forgive that kind of infrequent performance hiccup.
But rather, regular usage, especially of Confluence and Jira, just feels sluggish when doing bread and butter actions - adding a comment to a ticket or reloading a kanban view, loading a new page or opening the search dialog on Confluence.
Yet, everything I've heard about the responsiveness of the 'cloud' offering is that it drives people to move to the on-prem server version.
[+] [-] garenp|5 years ago|reply
It's hard for me to imagine that atlassian would allow individual instances of jira/confluence/etc in their cloud versions to get nearly as much resources as you need to give them on-prem. The incentive would always be for them to try to minimize the amount of memory every instance gets, which does NOT maximize your individual performance. Shared resources will also make performance more variable and "noisy", making it more difficult to narrow down specific causes. It's hard for me to see how "Java" and "cloud" go together.
[+] [-] smcl|5 years ago|reply
Oh and they keep updating the UI which involves shuffling things around and every new version is slower than the last. If you're running on-prem you at least get to decide when to update and notify everyone accordingly. We'd just find out one day that everything has been moved around and learn the new locations of different features.
Sorry I'm really just generally ranting about Jira now but honestly I find it so infuriating to use. Confluence as well, how did they manage to make wikis suck so bad?
[+] [-] yjftsjthsd-h|5 years ago|reply
[+] [-] tedivm|5 years ago|reply
[+] [-] bootlooped|5 years ago|reply
[+] [-] originalvichy|5 years ago|reply
[+] [-] pc86|5 years ago|reply
I've heard the same about cloud performance, which seems to be an argument against it being simply configuration, but who really knows?
[+] [-] ashtonkem|5 years ago|reply
Confluence just sucks though.
[+] [-] Spivak|5 years ago|reply
* It's really hard to scale support. You end up being on the hook for why "it's not working" in thousands of different of environments you have zero control over. Anyone qualified to do that level of support could be making more money not hating their life.
* Nobody wants to pay for updates. I mean it's totally rational but we also don't want to support $old_version forever.
* Even when updates were free people still didn't do it. Telling your users "sorry you're using an ancient version, please update" just makes people angry. Again, understandable, but surprise! those same people suddenly weren't angry about updates once they were on our SaaS offering. Turns out that frog boiling is reasonably effective.
* We had to deal with swaths of customer complaints about performance that we're mostly not our fault. Sorry our app is slow on 1/64th of a used SuperMicro. Maybe your IT department should get on GoFundMe?
[+] [-] zamp7less|5 years ago|reply
[+] [-] laurent92|5 years ago|reply
- Customer must start with cloud,
- Data Center is the overweight offering, starts with thousands of users, grossly overpriced like GitHub Enterprise.
[+] [-] jeff_redradish|5 years ago|reply
https://chat.goodbyeserver.org/
It's an difficult position Atlassian are inflicting on tens of thousands of customers. Atlassian's self-hosted products are uniquely flexible, being built on a plugin architecture, and many orgs have indeed customized Jira extensively with plugins, notably ScriptRunner [1]. Atlassian's Cloud plugins have their APIs, but have nothing like the same flexibility. A lot of functionality just isn't possible in the Cloud architecture. It's a bit like Firefox moving from XUL to an extension API.
In my opinion, the most customer-respecting way forward would be for Atlassian to open-source their discontinued Server product line. Go to the "open core" model with the clustered Data Center product as the upsell. This avoids screwing over their customers, and if their Cloud product really is as good as they say, customers will migrate to it naturally over time. It's the kind of damn-the-torpedoes move I think Mike CB would like.
But over the next 3 years, lots of painful migrating or evaluating-of-alternatives will need to happen, and perhaps a non-Atlassian forum for sharing experiences will help.
[1] https://marketplace.atlassian.com/apps/6820/scriptrunner-for...
[+] [-] thu2111|5 years ago|reply
[+] [-] chii|5 years ago|reply
little did you know, the source code for atlassian server products is already available for any paying customers. It's not FOSS, but if you need to run this yourself, and customize it, it's already possible. But of course, you will be responsible for keeping it updated (patching vulnerabilities from libraries etc).
[+] [-] ashtonkem|5 years ago|reply
A lot of what makes Jira so broken is that it is way too flexible, and this flexibility has costs in terms of performance and stability. There’s a damn good reason why I used Chrome back in the XUL days for FF; because XUL FF was not performant or stable for me back then.
[+] [-] benhurmarcel|5 years ago|reply
[+] [-] RoyTyrell|5 years ago|reply
[+] [-] pldr1234|5 years ago|reply
[+] [-] manicdee|5 years ago|reply
At least they if we mess up security on a host inside our business it’s not the end of the world.
edit: I don’t make the money decisions here, I am just a peon.
[+] [-] tupputuppu|5 years ago|reply
[+] [-] i386|5 years ago|reply
[+] [-] ameshkov|5 years ago|reply
Server products is what made us purchase Atlassian software. There are numerous reasons why our company (and a lot of other companies I presume) would like to avoid cloud. And if we really had to go cloud, I simply don't see why we would choose Atlassian over alternatives.
[+] [-] alelefant|5 years ago|reply
[+] [-] thijsr|5 years ago|reply
[+] [-] justin_oaks|5 years ago|reply
For Bamboo, I immediately think of Jenkins, although it's showing its age now.
For Confluence, I think of Media Wiki.
For BitBucket, I think of self-hosted GitLab.
Not sure about Jira or Service Desk, though.
I'd appreciate any better suggestions.
[+] [-] dagmx|5 years ago|reply
https://fee.org/articles/australia-s-unprecedented-encryptio...
[+] [-] joejerryronnie|5 years ago|reply
[+] [-] Waterluvian|5 years ago|reply
Atlassian feels like a bunch of individually purchased products that they then just jam together with integrations that never quite work the way they should.
I wish my company would abandon it.
[+] [-] elliotpage|5 years ago|reply
To be slightly more charitable, this is a sensible move as at least this way Atlassian can control the end user experience and (perhaps!) have it not be utter trash. Far easier for them to manage this way.
[+] [-] justin_oaks|5 years ago|reply
One thing I like about the server versions is the ability to spin up development versions of these applications so I can test out new plugins, version updates, or complicated configuration changes without affecting our production applications.
So much for that. I guess I can skip that step and when it goes wrong in production I can be the old man who yells at "the cloud".
[+] [-] originalvichy|5 years ago|reply
If only they made the DC cheaper I wouldn’t be so disappointed.
[+] [-] Fumtumi|5 years ago|reply
I'm using it for ages now and i have not seen any real features being added in the last 10 years and fundamental issues have not been adressed.
The weirdest thing was their UI change a few years back where they thought it would be great to make core workflows 1 or 2 clicks further away than before.
Like 'creating a ticket; went behind this plus button thing.
Plenty of features i would love to see, the community votes them up as well but nothing happens.
Apparently they are to busy with stuff no one is seeing. Hope that actually brings in innovation.
[+] [-] pmlnr|5 years ago|reply
Then there's fossil[^2], which is brilliant, and is completely self-contained, with version controlling, wiki, etc. It's a bit awkward, but nonetheless feature full.
[^1]: https://gitea.io/
[^2]: https://fossil-scm.org/home/doc/trunk/www/index.wiki
[+] [-] hnruss|5 years ago|reply
It would be nice if Atlassian open-sourced Bamboo so that we could continue using it long-term, but honestly one of the primary reasons that we’ve stuck with them is their excellent support team. Without them, our development team would be forced to diagnose and fix issues in Bamboo, which is not the best use of dev time. So I guess we’ll attempt to find alternative build software soon.
[+] [-] InsaneOstrich|5 years ago|reply
[+] [-] lenkite|5 years ago|reply