To all Atlassian server champions - we want to hear from you

[Thomas Dörfler]

Today I had a few minutes to rescan this forum and the messages. I tried to look at Atlassians enthusiasm for its cloud technology.

From a technical point of view, Atlassian might be right: Having a central, professional administration for the valuable data we all put into our Jira instances, having constant improvement, added functionality is ... cool. yes.

But from a legal/contractual point of view, moving to Atlassian's cloud requires constant monitoring: It starts with "Atlassian Data Processing Addendum": This Addendum points to a list of subprocessors in Australia, Ireland, Germany, the US. And this list may (=will!) change through time. When it changes, you can reject the change by closing down your Jira cloud instance. Additionally, when legal framework changes (like invalidating the privacy shield... what will the next 10 years bring in international data law?) you have to reinvestigate: Are we still ... safe?

So from my point of view, moving to Atlassian's cloud will make our server admins obsolete (no, not really ;-) ). But we have to add a few IT lawyers to our staff, which will monitor. And possibly enforce us to leave the Atlassian cloud.

Wow, what a change.

And, BTW: in the beginning we didn't want to enter the cloud anyway!

[Julius Zatroch]

Cutting off small customers is very bad idea. They are generating small amound of money but huge market potential and new customers in future. And im not talking about Atlassian partners too....   Look here what happened to another IT company LOGMEIN,  https://finance.yahoo.com/quote/LOGM/ .  They cut off small userbase in 2018, we have been forced to find alternative and after 2 years is stock price of LOGMEIN only 1/2.....

[a.stucki@solcept.ch]

Cloud will be a absolute No-Go for us for the following reasons (and to @Cameron Deatsch : they are not nice to haves of some random company policy):

• no version control (in functional safety projects we do "Tool Qualification" for each version of any tool we use, if we have no control on the version because some cloud god can change the version at any time -> no good)
• data retention for many years (the 75 demanded by an aviation customer might be too optimistic, but you get the idea...)
• data protection (at the moment there is no technical (root access of cloud admins) or legal (EU laws, not even pinning works) solution on the "legal" side, in addition it is difficult to convince security-minded customers of the security of cloud solutions -> no good)

I see this as an opportunity for a nifty competitor of Atlassian to catch a nice bit of the market ;-)

[Julius Zatroch]

Hello @Cameron Deatsch  and others Atlassians. As is your own status on Atlassian charity donations, you have 61000+ 10$ licenses now. What about to increase price to 100$  ?  I think nearly everybody can pay it per year. It will give you more like 6000000 $....  Should be enough to leave server line alive.

[Dalectric]

Can I point out that the timing of the upgrade is terrible for us in the UK (and probably many other countries in the world). We have budgets set out from April to March so don't have scope to change that until after March 2021.

If we upgrade before Feb 2021 we can take advantage of lower prices, but the cost increase for our Confluence server (3x current cost) is well above what we have as a contingency. If we upgrade after Feb 2021 the price increase is 5x higher.

This also doesn't take into account the current financial climate where we are under enormous pressure to minimise our costs.

I implore Atlassian to at least delay this decision by another year to allow us to budget accordingly.

[Vicky Blease]

Hear, hear!

[Yoke Pean Thye]

We are a volunteer-driven non-profit organisation that is active in three countries with volunteers from around the world, and works remotely. We have benefited greatly from your Community License; using Confluence and Jira has enabled us to collaborate much more effectively to achieve our mission. While we would be happy to pay for such a useful product, we can't afford much and even the Community License discount for your cloud products is far beyond what we are able to pay.

I sincerely hope you will do something to help organisations like ours.

[Jörg Godau]

@Cameron Deatsch 

It's all basically been said. I've personally been arguing for true GDPR compliance and full data residency for years.

We love cloud, we can't go there if you take any data (users, logging really any data) out of our jurisdiction. This applies to most of everything in but especially public sector and health care in Germany.

Yes we use some other cloud products - ones that provide such certainty around data regionality and GDPR. I've been begging for official announcements about this in slack and to everyone from Atlassian who will listen.

People invested heavily, even after the price hikes last year. Now this? It is breaking trust and breaking a guiding principle of Atlassian to not f@rk the customer. Just be honest and swap that one with "cloud first" and you'll be okay.

The big question now really now is:

How long before Data Center ist also EOL? How long does Atlassian guarantee it will maintain and upgrade data centre (for those that can afford it)?

Goodnight and good luck everyone

Jack...

PS. I am in favour of cloud, I really am, iff it is legally ok for me to use it. Features are great but come much later in the list.of must haves.

[Cameron Deatsch]

Jorg, thanks for your feedback. Data Residency continues to be a constant trend in customer feedback. I'm sure we will be able to achieve the requirements you have in these areas and will continue to push the product teams to deliver here.

As far as the long-term support for Data Center, we have no published timeline but we expect with this announcement we will see many customers move to Data Center and will continue to support them for many years. 

The best validation I can give on our commitment to Data Center is our public roadmap for Data Center: https://www.atlassian.com/roadmap/data-center

[Jörg Godau]

The question is when around data residency. I've heard you're working on it, but I haven't heard hard dates. If you know it is so important, then why not solve it before this move?

Unfortunately @Cameron Deatsch that probably won't cut it in terms of Data Center EOL.

At previous events, team tour etc, as recently as last year Atlassian stated publicly they were cloud first but keeping Server.

That statement/promise was broken. Data Center will be a huge investment for those that can even afford it, those companies have a right to know it's a five, ten or fifteen year horizon.

I would commend you to make the promise and back it up with full refunds if you don't keep it. As a public company, it seems only the money is talking these days.

[Seweryn Szatkowski]

...

[Julius Zatroch]

Ist unbelievable, how 18 years of building trust to customers are gone in 1 day......  

[Seweryn Szatkowski]

...

[David Goldstein]

Many Atlassian server customers—in considering a move to Atlassian's Cloud—don't initially realize that not only are the feature sets often diminished in Cloud apps vs. server apps due to Atlassian framework limitations, but each 3rd party Cloud app actually runs on that vendor's own hardware (typically AWS, Azure, Google)—entirely separate from, and outside of, the Atlassian Cloud.

Cloud customers not only need to rely upon Atlassian trust—and any data residency offered in high-priced, specialty service tiers—but need to separately vet and trust the security model, architecture, business stability and data residency of every Cloud app vendor that transmits data to, or stores customer data on, that vendor's own servers.

Perhaps the new Forge Cloud app framework is supposed to solve this 3rd party data storage issue. If not, what's Atlassian's roadmap for addressing issues of data residency and related security and privacy compliance for the 3rd party apps customers want installed on Cloud?

[Александр Грунин]

Sorry for my English.

We also have a 5000+ (JIRA, Confluence) users and cant move to cloud due to compliance reasons. Politics :(

I new a lot of organisaions in our country (Russia) wich cant move to cloud with same reason (gov, med, a part of education...). And DC prices also is inaccseptable with our budgets. :(

Currently in our forums, chats and other a lot of discussions about alternative solutions to Atlassian product.

[Dwight Holman]

Since Atlassian's announcement (some time ago now), I have worked to determine if migrating to Atlassian Cloud products is a viable option for our organization. This has taken time, and I think we have explored this reasonably thoroughly.

Our current conclusion is that it is not practical to migrate from Server to Cloud because:

• Many of the apps we currently use on Server have no equivalent on Cloud.
• The performance of Cloud is slower than Server.
• The migration process is time-consuming and painful.
• The ongoing operational (support/subscription) cost would be about double our current cost (even if it was possible). The cost calculator here is rather comical in my view: https://www.atlassian.com/migration/cloud/savings-calculator because it makes no consideration of 3rd party app costs, that additional costs may be required (e.g. Atlassian Access), assumes that Server maintenance is always done during office hours, and that JIRA Server runs on dedicated hardware.
• We would have to accept that our customer's and company data would be stored in a data centre outside of our country.

As others have already noted - this decision seems to be in direct conflict with Atlassian 'policy' to not screw the customer. There is no current reason we would migrate to Cloud.

The most logical path seems to be; continue using JIRA Server as long as it keeps running, and stop paying for support at the EOL. Whilst simultaneously hoping we can find and migrate to a product from another vendor.

We are a small organisation and DataCentre is cost prohibitive.

If that's how it plays out, it seems Atlassian would lose the most as a result of their announcement. How do the shareholders feel about the plan I wonder...

[Tobias Ravenstein]

The calculator also asumes that the developers can not work if jira is down which is ridiculous.

For the Cloud products you get a SLA of 99.9% uptime ( https://confluence.atlassian.com/cloud/service-level-agreement-for-cloud-products-973486377.html ) which means the service can be down about 43 minutes every month without compensation which could cost you 36$ per user/per month (by the same ridiculous asumption that developers can not work during downtime) - or eat up you savings. And you only get the guarantee on the premium plan.

[Thomas Dörfler]

From our point of view, the effort to keep track with the legal stuff is also considerable. Atlassian can change its terms of use and its addon provider locations from one day to the other. We must keep track wither relocation and change of contract is ok. If not, we have to save data, pull them from the cloud and shut down our instance.

On our private Server, things are MUCH easier to check: Once during installation, that's it.

So in fact we would move support from our IT to our legal department (oh wait, we do not yet have one...)

[Heide]

Sadly, Atlassian sends the message, that they do not care about their small- to mid-size customers (far below 500 users) in industry sectors which cannot use cloud-based solutions due to regulations/security issues/... 

[gzgenm]

due to insane increase in the price LOL

[Diego Bruno]

Atlassian became Atla$$ian .. Now it prioritizes ($$) over Customers and ($$) over, Consultants and Developers thanks to whom Atla$$ian grew around the world.

Atlassian lost its essence and disappointed me

Atlassian let me down :'( 

[Tom Shaffer]

I've got the feeling that after a lot of us vote with our $$, we'll get a much more affordable option being offered to us all in the name of "you spoke and we listened." (Blech)

[Stephen Sifers]

@Diego Bruno,

Reaching out to ensure you're aware of the Rules of engagement for the Community. We've noticed you've posted this same message at least 6 other times across Community. With this said, we do value and listen to feedback, including constructive and critical feedback, however, spamming the community with the same message across multiple posts does a disservice to those expressing their actual impact and concern to Atlassian.

We'll be removing your duplicate posts and leaving this original post as-is to ensure your message is heard. We would suggest reviewing the above rules of engagement and our Atlassian Community voice and tone guide.

Regards,
Stephen Sifers

[Diego Bruno]

Thanks Stepehn, I will follow your recommendation and keep my mouth shut.

It seems that a i cant express my disagreement... it smells like some kind of censorship... 

by the way, I never did copy an paste ... I just answered different posts in a different way but expressing my disappointment with Atlassian.

Regards...

Diego

[Vicky Blease]

Seems Atlassian are adopting Trumpian tactics and deleting posts they don't approve of.

Classy,

[Stephen Sifers]

Hello Vicky,

I can assure you we're not deleting posts, most of the time posts become caught in our spam filter and it requires manual interaction to take them out of spam. I checked your profile and am not seeing any removed posts and no items caught in spam currently.

Please let us know if you're still missing content or a post of yours is not visible.

Respectfully,
Stephen SIfers

[christophecariou]

Hi,

Most of our atlassian products deployments are inside internal private networks. By this way, non authenticated users can have read only access to company knowledge, from Jira, Confluence, Bitbucket...

How can we achieve these use cases in SaaS architecture ?

Regards

[Julius Zatroch]

I think Atlassian dont understand what's "Enterprise cloud". Only available option for us is to create own SERVER software instance in Azure, AWS or Google, secure it with OWN software firewall and connect VPNs to company infrastructure. We cannot depent on Atlassian in terms of security. Look on this table:  https://stack.watch/product/atlassian/jira/   Every year higher numbers of JIRA vulnerabilities - 45 this year.....

Lets add server versions to AWS Marketplace and MAYBE we can think about transition to cloud, but "our cloud"... 

[Mike Rathwell]

@Julius Zatroch that determination is precisely where I am right now. As much as the lazy human in me would like it if I could go cloud (both privacy and technically), I may well yet end up building out a shiny new AWS DC environment in place of my existing AWS server environment.

If it ends up being the DC route, since my DB is already multi-AZ, I am facing only a move from EBS to EFS, licensing, and adding nodes as I see fit.

Gosh... come to think of it, that seems MUCH easier :)

[Mike Rathwell]

I saw this question posted and (somewhat) purposefully sat it out a bit to see the prevailing view to this. The other reason was to think about how to answer the question as this is, most definitely, bigger than a breadbox. This is likely going to be a rather long post (long enough that I wrote it in Google Docs to guard against losing it in case Something Bad Happens).

So… Where to start…

As I looked through all the various posts, it seemed like it boils down to a couple of things; cost and function.

Of the two, the one that might be the easiest to dispense with is cost.

In my world, will this cost me more? Yup. However, I have been and am at a place that spends far more on things that, to my tiny mind, return rather less than even what the increased cost of the New World Order will be. The increased cost, for me, won’t necessarily be a Thing as going cloud has been and continues to be “suggested” and the extra spend will not be unexpected. For those where annual spend is more of a concern, I do note that Atlassian is paying attention to this even if they aren’t necessarily saying “Ok… we will give you all your things back. My bad.”, and perhaps this feedback would filter up to address this. None of us are being expected to assume the higher cost immediately and there is a good grace period.

When looking at function, we are now in the rather larger part of this comment. To give you a bit of background on myself to help understand why my thinking is what it is.

In my <mumble> years of being in the technical world, I have been a system administrator, designed control systems for utility power stations, designed large scale biometric systems for government and law enforcement, been senior enough to manage an entire data center and its denizens for a rather large bank, and yes, be an Atlassian Administrator. I am at a stage in my career where I have studiously worked myself back to being an Individual Contributor but considered “leadership”. The work I do has me being sought for how to do a <insert big thing> for the company and my explicit role as Atlassian Administrator lets me actually deliver some of that.

When I have been in a system administration role, I was one of those that worked in a wide variety of UNIX environments on hardware that only those of my <mumble> generation will even have heard of. Add in Linux, a tenure with super super privileges on an entire bank of Tandem Nonstop systems, and now containerization on AWS where I containerized Jira and Confluence before it was “cool”. A lot of the ability to leap around also comes from my perhaps perverse enjoyment of “shiny”. 

Why did I present all this minutia about myself? Because, as an engineer, in the classic academic sense of matriculating as an engineer and not just an appellation, one overriding approach taken previously and will work in this context is:

Go back to first principles and solve. This <thing> is made to do this <function>. It’s not where it was on <previous thing> and it maybe does it a different way but I am paid the big money to figure out how to make it do this <function>.

All of this brings us to the topic at hand: Some Big Things just changed and it’s now time to do something about it.

Cost aside, many of the comments appear to be in the “things are changing and I don’t like it” vein. While understandable and valid, things do change. If I were looking for analogs outside this discussion a couple of notable ones (among many) leap to mind. When Apple stopped putting headphone jacks on iPhones, the entire world lost their minds. Apple is still selling lots of iPhones. When Microsoft release Windows 8, people lost their minds because no Start button. Microsoft caved and basically added a Start button to Windows 8. I was sad to see that. I was an early 8 adopter at work so it had to work and loved it. Microsoft had what I found to be a real environment that handled both tablet and desktop functions well and could be both as one moved from place to place. I was sad to see that capitulation but Windows 10 is still the predominant desktop OS.

A “non-imposed” change I tried to effect was a spectacularly successful POC but fell flat for purely emotional reasons. I noticed that my working life was largely on a browser so… was there a way to live day-to-day on ChromeOS? The answer turned out to be yes as proven by two years of exclusive ChromeOS as my daily driver. In my case, even as a system administrator, only small alterations were needed to do so. I was able to have a powerful Chromebox to use at my desk and also a Chromebook should I need to venture away. All the things were always there and at a cost about ⅔ the cost of a typical laptop provided. Lower maintenance effort, good central management and little chance of data loss. Would anybody move? No.

At my shop, I have heard the “Why aren’t we on Cloud” refrain for a few years given the (apparent) mindset of “SaaS All The Things”. Periodically, I would do a review to see if I could be cloudy primarily to address the refrain but secondarily because I am not averse to it. While we are not a large in numbers Atlassian shop, we are huge in diverse functionality and I am the only Atlassian Administrator. Along with the usual technical suspects, basically nothing moves here without a Jira issue attached including in Marketing, Content, Design, Creative, Social, Legal, all facets of Finance, and HR. Each of them have their own needs and I strive to make Jira and Confluence feel like “home” to them which has led to a very high and happy adoption rate. Our Legal department, to a person, thinks that every legal firm should run on Jira and our Marketing PgM commented that she “couldn’t be successful without my tools” and “I got my lunch hour back”. With that level of complexity and also being a full stack engineer that has built out the environment from an empty AWS environment up to all that, not having to look after the operating environment and keeping things up to date has its attractions.

Until recently, the assessment of going to the cloud has been bounced very quickly on purely technical assessments. For an environment as customized and tuned as ours, there simply wasn’t enough function available in the cloud, mostly a case of key apps simply not there at all.

This time it’s different but not because of the recent announcement.

In between formal assessments of whether to go cloudy or not, I keep an eye on what of my “needs” have become wispy, as it were. Since I did my last assessment, enough items are “there” now that I am looking at it from a different perspective. While before a review took a matter of days and closed because X number of key items weren’t there, X has gotten small enough that a deeper dive is needed. As such, I started with a legal review of the ToS quite some time before this ostensibly contentious announcement. If they bounce it, then I have my answer and direction; on prem. If they don’t, then this deeper technical dive is in order.

To be very clear, I am not expecting to come in and make a pass/fail on a 100% like-for-like set of functionalities. For the “misses”, I will be taking a “first principles and solve” assessment of “can I do <this thing> a different way?” approach. If it looks like I can, even where there is a materially different way for <this thing>, my user base trusts me and will work with me to make it happen.

My next steps will be a series of decision gates.

The first gate is whether legal passes the Cloud ToS. If they do, then I do the deeper dive into Cloud to see if I can deliver the function needed to my user base. The second gate is whether the deeper dive says I can, indeed, deliver the functionality needed in Cloud or not. If I can, then the heavy lift to migrate will start. If it seems I can only provide expected function on-prem, the third and final gate is to pragmatically decide if I move to DC or ride the less expensive server train as far as Atlassian will let it go.

Assuming I do end up staying on-prem, I am not likely to remain on Server Edition regardless. The number of DC apps I definitely need are all there now. Migration to DC would be relatively simple as I could lazy out and start with just running DC single node. Make more new stuff later. That said, while my reliability has been stellar, it figuratively keeps me up at night worrying about AWS losing the AZ my docker host is running in. I do have the production database configured multi-AZ and I do have a tested mitigation plan but I would rather not do that. Given that Jira and Confluence are literally mission critical to the shop, multi-AZ resilience is important. Less important in my shop, but shiny, is the ability to spin nodes as load comes on when needed. As such, I was fortunate to have already primed leadership that the Atlassian stack was going to cost more as I approach maintenance renewal next May even before this announcement.

Given all the above, I had already started the conversation on going one of the two avenues available in the Atlassian announcement, even though we are a relatively small shop, based on sound reasons. As I move forward on this journey to a destination either among the clouds or on the firmament, along with the engineering maxim of “go back to first principles and solve”, I will also follow my other approach best typified by a quote by Robert A. Heinlein: “Always listen to experts. They'll tell you what can't be done, and why. Then do it.”

[Dalectric]

An interesting post, but I don't agree with all of it, for a start I've seen very little of people complaining that something is changing and we don't like it. People are complaining that the changes have a significant impact on their business and use case.

From my review of the posts and comments on these forums I'm seeing the following common themes

1. We cannot move to cloud because we need to have our data on prem
2. For those that need to stay on prem the cost increase to move to data centre is prohibitive and there is not enough time to plan a budget accordingly.
3. Cloud doesn't offer the functionality we need, or allow control of the application to prevent functionality being lost/broken

[Thomas Dörfler]

Mike,

"go back to first principles and solve"

surely this is a sound principle. Funny enough, we were at that point two years ago. Due to our samll company growing we found we need a better way to steer projects than "let's sit together and see what to do next".

Two years ago, we roughly evaluated some software packages which looked promising. Due to its reputation, we jumped on Jira, a decision we were happy with. We invested a fair amount of time and money to really get productive with this system. We also had to overcome some personal dislikes, but that's always to be expected.

Now two big things have changed for us with Jira:

- The future outlook: we really expected to have a stable, reliable, evolving(!!!) basis for our business, not something with sharp edges and a revolution.

- Security: We store most of our knowledge in Jira, and part of our customers' knowledge (aside from personal data, which is its own issue in Europe). We decided that we can trust such a system, when it is located on our own private server, behind our firewall and with some monitoring. When moving to cloud, we understand that our data currently will be moving around the world to be processed by add-ons. That's ... technically cool, but untrustworthy from a security point of view. (I know, that Atlassian states to work on this... when will we see that done?)

So, back to your statement: "go back to first principles and solve". We will wait for 2-3 months, maybe our complaints here make Atlassian ... think, learn and reevaluate (I really would love it!). And then we will go back and solve: Look for the SW system, that fits our needs.

A cloud based Jira will most likely drop off the list rather soon. Unfortunately...

[Mike Rathwell]

To both @Thomas Dörfler and @Dalectric , I too am in the state where I am determining first whether I can go off-prem with my data (figuratively; I already host my on-prem server on AWS). That was the legal question I have in front of our attorneys at the moment. Only after THAT do I start to do the deeper technical dive to determine if I even can be cloudy.

I am jumping on this early because I had a Cloud/DC move proposed already with the one pager initial budget chased up the chain. I do agree with all that this is, eventually, going to be a big nut to swallow; Cloud/DC are loosely the same cost to me and that cost is ~2X what I pay now for maintenance renewals under cloud.

I could be reading the announcement wrong but, aside from the hard stop on new licenses next weekend, we can still pay our current support for another 3 years.

I did look long and hard at that 3 year thing and the caveat that concerns me about taking that route (and haven't seen it spelled out) is if the stop of server license sales also equals a stop of server add-ons. My environment is pretty stable in that regard but about once a year someone comes to me with an OMG this is critical need that, to deliver, sometimes takes a new add on. 

I do agree with you that the price increase, especially for smaller and/or non-profits is onerous. I am fortunate in that a planned cost bump to go cloudy or DC was in the cards and I am a size that fits the current license model. I also agree with you that I would hope that Atlassian would look at the pricing structure even if it means just extending the qualifications for Academic level pricing along with some of the smaller tiers.

In the few times I've had a chance to talk to the PdMs, I have commented that "small" does not equal a simplistic system. There does seem to be an underlying assumption that a smaller environment is also simpler which, aside from data privacy concerns, would be just fine on Cloud. That is not always a true statement.

[Mike Rathwell]

Update 10/29/20: Legal have just given a provisional "approval" needing only InfoSec blessing now that, should I find a way to do it, I could move cloudy.

More on this festival as it progresses.

[ITMAGE]

To complement what I wrote earlier (which now seems to have mysteriously disappeared into the abyss), I just noticed that the 'Data Center' license is an ANNUAL payment rather than perpetual license. This new realization makes it completely 100% out of reach for us for the foreseeable future. Given the current pricing of DC, once we grow beyond our current license count, we will need to seek alternatives.

[Erick van Rijk]

Is this thread locked or something? I spend time to write posts but it goes into the void...

edit: I guess not. So I wrote to different post and both are gone. (even in my own "posts" section). Anyone know what's going on?

[Erick van Rijk]

@Stephen SifersCould you check my post for spam as well, since I lost 2 posts. One in edit and one regular

[Stephen Sifers]

Hello Erick,

I checked your post history and I am not seeing anything within spam. I do see multiple posts from you since the 17th, and nothing is listed as spam currently. To check your posts, navigate to your avatar in the top right, and select View Profile (All of your activity will be listed there). If an item was marked as spam it will show "In Moderation" pending an Atlassian to take action.

Respectfully,
Stephen Sifers

[Erick van Rijk]

Hi @Stephen Sifers Thank you for checking.

I did check the my own posts section but didn't see anything there. I probably made my point by now, so lets leave it at that.

[Kyle Hardin]

I'm most concerned that the death of the Atlassian Server product seems to be the death of the on-prem nonprofit licensing program as well.  I know that Atlassian wants to push to cloud, but many nonprofits will not be able to utilize the cloud for a variety of reasons, often related to their work.  Their only option will be Datacenter, which has no nonprofit licensing and requires a 500 license minimum purchase and annual subscription.  

We serve remote Alaska and have a WAN connection of about 4mbps with a low bandwidth cap. Cloud is 100 percent not an option for us given our current WAN options and intermittent connectivity.  The discontinuation of Atlassian's on-prem nonprofit licensing program will be crippling for our on-prem environment and the services we provide to the community.  

[Alex Janes]

@Kyle Hardin We are also in the same boat. We are a non-profit that handles medical data, and therefor we have HIPAA compliance requirements. Their cloud just isn't able to handle that right now.

The only thing keeping me going right now is that @Cameron Deatsch mentioned in another thread that they are going to be updating their community licensing plan. No details yet. But hopefully it just shifts the nonprofit program to datacenter?

[Kyle Hardin]

Thank you Alex, I appreciate that additional perspective.  I truly hope that Atlassian does the obvious thing and shifts the on-prem nonprofit program to Datacenter, but I honestly expect that they'll choose to remain bullish on Cloud and focus on wrapping up the compliance hurdles there instead, ignoring the connectivity issues plaguing more remote regions.  

[Patrick Löffler]

Hope this is a bad joke.

When will the Atlassian Cloud be TISAX certified?

Without that it is a No Go for us.

[Andy Heinzer]

Hi Patrick,

Even with these changes we just announced, you can still continue to use server for the next three years and you have plenty of time to explore your options for the long-term. If you are still unable to migrate to cloud by the end of the three-year support window, you can move to Data Center. Data Center will remain our self-managed enterprise solution that enables you to maintain control over their environment, helping you meet your unique security, compliance, and scale demands.

We’ve built a number of tools and resources to help you navigate this change so please check them out on our website.

We also have a tracking ticket (CLOUD-10934) that you can follow for updates on the availability of TISAX compliance in cloud, as well as our cloud roadmap.

[ASTEN RATHBUN]

This is incredibly tone deaf.  

For customers with multiple networks in the 20-50 user range, data center is mind-bogglingly expensive.  So, we are #$@!ed.

We can't see into the future nor have funding to buy new licenses for the next three years by February.  So, we are #$@!ed.

I never, ever want to see Atlassian claim "Don't #$@! the customer" is part of their values again.

[Taylor Huston]

@Andy Heinzer "Data Center will remain our self-managed enterprise solution that enables you to maintain control over their environment, helping you meet your unique security, compliance, and scale demands."

If you can afford it.

[Dalectric]

@Andy Heinzer this ignores the fact that we are financially penalised if we don't upgrade from Server to Data Centre before Feb 21 and the longer we leave the upgrade the more expensive it is for us. And the price difference is not insignificant.

[David Willson]

It's interesting that "Dont F* the Customer" now needs to be interpreted. Part of the beauty of this value, is that on its face, it seems so self evident. But perspective matters, and now something so simple needs interpretation.

I've been wondering about this since the announcement. In a sense Atlassian isn't F*ing the customer. They have been open and transparent that cloud is the future for their company, they have said "things are changing" and "here are your options" oh and "here is a long runway to make your decisions and a bunch of free resources" to help you change however you end up deciding to change.  When you consider it from Atlassian's perspective, with the inevitability of cloud, change is just coming and from their perspective they have gone out of their way to communicate and help customers transition.

From the customers perspective, Atlassian is forcing a change which has a cost. Think about just the amount of time people have spent talking about the change and posting on this forum. That has cost companies. Then if you go to cloud, there is planning and architect and migration costs plus new subscription costs. Or if you 'change' to datacenter - which is arguably the simplest solution to the announcement - a large # of customers will have a 4x+ increase in license costs for an operationally identical product. If you are one of these customers you are feeling INCREDIBLY F*ed.

So as a customer, and a community leader, it is such a feel bad. And if Atlassian was like any other faceless organization and didn't create so much high quality soft skill value driven messaging, I probably wouldn't care. But it feels like a betrayal of their values. I bought into their message and so now I feel physically sick and betrayed with this direction. Shame on me for listening, and caring so deeply and believing, and shame on Atlassian for forcing everyone in the ecosystem to now have to interpret one of their most simplest core values.

The solution is to provide a lower tier of DC. I wrote a open letter (link below) arguing for this.  In the bigger sense Atlassian's reputation is at stake for all of us 'forgotten middle' customers. I would really like to get the letter to the founders so they can hear what making dc affordable means. If you have a means of sending it to them, directly or indirectly, and could help me out that would be much appreciated.

https://community.atlassian.com/t5/Jira-articles/An-Open-Letter-Atlassian-s-Forgotten-Middle-re-The/ba-p/1510036

[Erick van Rijk]

So as a customer, and a community leader, it is such a feel bad. And if Atlassian was like any other faceless organization and didn't create so much high quality soft skill value driven messaging, I probably wouldn't care. But it feels like a betrayal of their values.

This is a move I would expect from Oracle or SAP.

[Nuno Gomes]

The solution is to provide a lower tier of DC.

Well, it already exists: it's the "server" product. It was shot down two weeks ago.

Think about it. What is DC if not "server" with "enterprise" (high-availability) features put on top?

[Dalectric]

Exactly....

[bbrks]

I appreciate my use-case is probably unique and I am very small fry, but I do want to voice my use-case, given I'd be willing to pay 20 times the amount I currently pay for a single-user, self-hosted Jira instance. $16/month consumer subscriptions are not unheard of, and it's something I'd be happy to pay for a self-hosted personal Jira.

I'm a developer that uses a single-user self-hosted Jira instance that is sometimes used offline (either hosted somewhere on the local network, or running on the machine I work on). This is super useful if travelling without network. Moving to the cloud does not work for me. I need it to be accessible and usable offline.

I use Jira for personal software projects, but also for project management (like tracking house renovations, gardening, etc.). Nothing I have found lets me link issues together like Jira does (Blocked By, etc.). In a large project like a house renovation, lots of things are dependent on another previous issue being done, and it's super helpful to get the bigger picture of project dependencies like this.

If anybody has suggestions for alternatives that use the "graph" like feature of linking issues together with constraints, please, I am completely open to any alternatives.

[Gianluca - STAGIL]

Hi @bbrks ,

I suggest you have a look at Stagil Asset - Advanced links

As the name suggests, you will be able to use custom fields to establish advanced links between issues, define hierarchies, and display them on direction/network graphs.

[Julius Zatroch]

Hi @bbrks 

Ping me and I will give you alternative on premise project software to try. They have even direct import of JIRA projects.... I will have tomorrow Zoom call with this company to discuss options and next steps...

[ASTEN RATHBUN]

For our organization, we cannot use cloud.  We have multiple small independent private networks.  Most of these are for very small teams, each with their own instance.  It's easy to spin up a VM and a dockerized Jira/Bitbucket/Confluence setup for small projects on each network.

If Atlassian takes this route, we will rip out all our Jira, Bitbucket, and Confluence.  I have no idea what we'll replace it with, but Cloud is simply impossible, and Data Center is hilariously not the product we need.  

[Dirk Ronsmans]

I'm curious how you manage your license then today? If you spin up an instance when needed do you also have multiple licenses that are unique or are these shared across instances?

[ASTEN RATHBUN]

Yes, we have a number of licenses.  Instances tend to stay around a long time, but are independent.  As projects wind down we remove old projects and use the same license/instance for new ones.  

[Cameron Deatsch]

Asten, thanks for the feedback. 

I see you work for a large enterprise organization. I assume Data Center doesn't work for you due to the federated deployment approach you have for teams where each team gets its own environment?

Our Cloud enterprise offering offers almost that exact approach you're looking for. Is the requirement to keep these systems on internal private networks due to security, apps, or something else?

If you do choose to go with another vendor would you be willing to have a call with me to discuss what direction you are going? 

[ASTEN RATHBUN]

I do, but our organization is kind of independent and we don't use the enterprise versions of your tools.  We can't.

Both customer requirements and security make anything but self-hosted impossible.

I have no idea what we'll go with.  We've invested so much into our configuration, our workflow, and integration that this is going to have a tremendous impact on us above and beyond the licensing.   We haven't even looked at other options in years as this was just perfect for us.   

[Thomas Dörfler]

Simple question to Atlassian:

if we would move our Jira server instance to your cloud: For how long do you guarantee access to our data? I am sure many customers need to maintain access to Jira data for 10+ years just to have access to project related information.

So: what is the guaranteed period you will maintain your cloud up and running?

[Viktor Kuzmychov]

Also what would happen to the data if subscription is not paid anymore? Server could be left alone with no support but with data accessible. Can the same be done to Cloud?

[Cameron Deatsch]

Thomas, we maintain your data for as long as you renew your subscription. If you choose to stop using the product you can export your data to comply with your various retention policies.

[Murty Chitti]

Due to country specific regulations we need to keep data within  the region like China or Europe.  Atlassian cloud will not provide this functionality.

We need this functionality for Jira ServiceDesk. Any alternatives?

[Cameron Deatsch]

Murty, we are working on our data residency capabilities and you can track these in our roadmaps here: 

https://www.atlassian.com/roadmap/cloud

We have 2 data centers in Europe in Germany and Ireland. We have yet to invest in a data center in China but definitely understand there is an opportunity there and in other countries. 

[Ferguson]

Atlassian is forcing us to move to the cloud. 

For now, customers can move to Data Center if they don't want to move to the Cloud.

But by looking into the Atlassian trend, Atlassian will stop Data Center support as well in the future and ask customers to switch the cloud.

In Cloud, you can't add additional security. It means for security you are totally depends on the Atlassian. Which is the thing I really don't like.

Customer like me for which Security is very important will face a many problem. 

Cloud security is tight, but it’s not infallible. Cybercriminals can get into those files, whether by guessing security questions or bypassing passwords.

But the bigger risk with cloud storage is privacy. Even if data isn’t stolen or published, it can still be viewed.

VERY VERY BAD DECISION. LOOKS LIKE ATLASSION IS THINKS ABOUT CUSTOMERS.

[Cameron Deatsch]

Hi Ferguson,

Atlassian spends a lot on security and has a sizeable security organization constantly looking for holes in our security programs. 

You can see more detail here: https://www.atlassian.com/trust/security

Are there specific additional security measures you would like to implement? Are there specific vendors we should look to build integrations with in this space?

[Thomas Dörfler]

Cameron,

I assume most your customers are working with constantly shifting requirements, also regarding data security. If you look back five years, many threads have shown up nobody was thinking of in the past. On the other side, legal requirements are on the move too, and they will differ for individual work areas, individual locations, individual customers.

For a server based product, each of your customers has to ensure, that security fits the customers requirements.

When we all move into your cloud, you will either have to ensure, that your cloud will fits the superset of ALL customers requirements, or you have to admit that at least some customers will not find the level/aspects of security they will need.

I don't see a plan at Atlassian to even collect the various requirements before announcing that the Atlassian Cloud is the perfect future for all. That's... a bit disgusting, ain't it?

[Tom Shaffer]

This whole thing has the vibe of - "We know our cloud offerings aren't as mature and well-built as our server ones but we promise we'll fix it some time in the future just for you if you sign up with us now."

[Dalectric]

Agreed. Atlassian hardly have a good track record of fixing issues with their server based products so I'd hardly trust them to fix issues with the cloud offerings.

You don't buy a product based on what it might do in the future, you buy it based on what it does right now.

[Stacey E. Kimmel-Smith]

My observation about this announcement is that the message is confusing, in part because of the range of products, add-ons, tiers, discounts, and exceptions that are not consistently referenced in the email or on the web site.  As an academic institution using 3+ Atlassian products, we need more clarity on our options and the pricing implications.

Will Atlassian offer Zoom or phone consults to help their customers work through their options and particularly, the pricing implications?

Thank you.

[Paul Stallworth]

Hi Stacey,

I'm also in higher education.  One thing we've found helpful is to engage a partner reseller to act on our behalf with Atlassian when it comes to licensing.  The partner can help with things like co-terming licenses and also navigate the complexity you mentioned.  This is something you can get for free, as the partners work it out with Atlassian.  If there are local partners in your area, they may be able to serve as your reseller.  If not, many of the biggest names in the ecosystem (Adaptavist, Appfire, etc.) do it as well.

Hope this helps,

Paul

[Cameron Deatsch]

Stacey, we have a team of customer advocates ready to help you through the confusion. Just reach out here:

https://www.atlassian.com/company/contact

If you don't get what you need, post here and we can set up a Zoom call to discuss.

[Tom Shaffer]

My observation about this announcement is that the message is confusing, in part because of the range of products, add-ons, tiers, discounts, and exceptions that are not consistently referenced in the email or on the web site. 

I completely agree. When I reached out to support about it last week (using the same link @Cameron Deatsch shared) I got a copy/paste answer of the exact same confusing/conflicting message from their website and my actual question was never addressed. Ah well.

[Stacey E. Kimmel-Smith]

I contacted Atlassian Monday to get a quote and to find out if there is any advantage to us renewing early to lock in current pricing. Three emails later and one referral, no answer. I requested a phone consult -- they nixed my first suggested date/time, I immediately replied with three more, and nothing since 10am Wed. 

And this is sales more than customer service which typically is the better experience. This level of response is more acceptable when we have local expertise and control. When we go to the cloud, and have far less control, I wonder what service we can expect.

Thanks for the information about the reseller.