I am curious as to why Atlassian chooses not to maintain their documentation sites in the current version of their own products and utilize Crowd for SSO? It seems that they may have a better understanding of the issues that their customers are complaining about in terms of usability, feature loss, critical feature issues, and upgrade problems if they actually did the upgrade first and be their own alpha/beta customer instead of having to deal with the aftermath of customer reported issues and complaints and understand the pain that their customers will experience.
I managed an engineering lab for a product and we utilized our own product at all times; consequently, when we worked on new releases, we were our own alpha customer and were able to identify and resolve the issues before they made it out. As daily users of our own product, it was clear the problems that we introduced, especially in terms of usability and upgrade processes, and thus they never reached our customers.
I don't understand why Atlassian doesn't invest in the same approach? If they used the products that they removed features for, they may understand why we complain about the request to reinstate such features as improvements, instead of bugs, or not deprecating them without a replacement feature, if it really was a feature being used and provided high-value.
They do dog food their download products all the time and have done so since the beginning. While they do have a few sites using OnDemand (like https://ecosystem.atlassian.net), they do not use it like a lot of customers would (no custom tabs for instance). So, they are not going to see the problems as much as customers would. I would prefer to see them roll out OnDemand changes to a smaller set of customers first to catch problems early so they can respond before everyone gets impacted. I least that is my 2 cents worth anyway :).
It doesn't seem like they keep their documentation and support sites utilizing the current versions, even if they are not OnDemand. Many issues I run into are not really on-demand - (e.g. issues regarding the new RTE, PDF exporter, and Editing in Office document features are not OnDemand specific). And, I don't understand why I have to have separate logins to all of these applications and they don't use Crowd.
It doesn't seem like they keep their documentation and support sites utilizing the current versions, even if they are not OnDemand
Which sites are these? We have our main doc site (confluence.atlassian.com) on Confluence 4.3.2 and support.atlassian.com on JIRA 5.1.8. Confluence 4.3.2 is presently the latest non-OnDemand release and JIRA 5.1.8 has not been released yet (we are dogfooding it).
We have lots of different servers running/dogfooding different scenarios, mission critical servers like:
jira.atlassian.com - JAC
confluence.atlassian.com - CAC
support.atlassian.com - SAC
Those are based on the behind the firewall (BTF / OnPremisses) versions of the products as they need to fulfil different requirements from the ones offered by the OnDemand platform solution.
We try to keep those up-to-date with the latest versions available to customers, although given it importance for the whole business (and some Atlassian behind the scenes integrations) it requires longer upgrade cycles and sometimes there are multiple new versions available to download before we can upgrade those servers.
Talking about OnDemand we do use it on a daily basis in more than 3 production-like servers to dogfood all the development that is done for OnDemand, in a exact match of what customers will see weeks in advance, In regards to ecosystem.atlassian.net as pointed by Bob Swift that one is already on the production-channel so doesn't contains the alpha/beta stuff, whereas the other 3+ are on a dev/dogfood channel and are only available for internal use/dogfooding currently, we may in the future create a beta-release channel where you could opt-in to experience new features before hand, although that is not in our roadmap for the time being.
In regards to dropping features, I as a developer and daily user of the product also felt a bit disappointed with the decisions, although from a product evolution direction it was crucial for us to make some of the changes as some of these features, while extremely useful for a small subset of our customer base, were impacting greatly on our ability to move faster and deliver bug fixes, improvements and new features quickly, as they were developed using an older/deprecated integration stack (from JIRA Studio days - 2008) between the applications and that is now superseded by the new header and a lot of behind the scenes new integration code.
I understand that you may simply don't care about this developer talk, and I'm fine with that, also while it's definitely not ideal to drop features without a proper replacement we are really trying to improve the product as a whole and the reason for asking to reinstate such features as improvements is that they will have to be re-developed on this new integration stack, and we can't just re-use the old code, if we could we wouldn't certainly dropped without warnings. The new implementation for these features when they come alive will enable customers to have the same feature working OnPremisses and OnDemand.
We have a project underway to streamline the customer experience in regards to login on all the different services we offer, but it's taking a bit longer than we anticipated due to some major roadblocks on establishing what is the unified identity for a customer across all sites. We do use Crowd in JAC/CAC/SAC but for historical reasons they were born and grew as separated installations which this project I mentioned before is trying to fix.
I am encouraged that you are delaying your unified login because functionality seems more important in this case. An user interface can be adjusted more easily than its underlying functionality. I am also encouraged that maybe some of the lost functionality in the 5.2 EAP may come back such as cross product search in some future release. I am discouraged that ease of use seems to get lost in Atlassian's decision making like losing the login gadget without a similar ease of use interface to replace it. It is my opinion it is better to delay releases if there going to loss of functionality or ease of use until Atlassian can adjust things. You do have the EAP program and the labs concept to gain user input and let people know what is going to happen. Yes, there are exceptions, but your customers will be much happier..
Atlassian Summit is an excellent opportunity for in-person support, training, and networking.Learn more
Connect with like-minded Atlassian users at free events near you!Find a group
Connect with like-minded Atlassian users at free events near you!
Unfortunately there are no AUG chapters near you at the moment.Start an AUG