Are plugins a feature or a bug?

Davet April 15, 2012

We have been using Confluence going on 5 years now and are now stuck fast at 3.5.7

Initially the holdup was with Customware scaffolding/reporting/composition late last year. This was fixed by paying hundreds of dollars (extra, every year:)

Now the holdup is Adaptavist with their "Essentials" pack licensing. They are using the same excuses as Customware i.e "those are not our paid plugins therefore they do not get priority!" I am expecting a pricing announcement any time now.

I can only assume that those of you who have successfully upgraded to Confluence 4.x have sensibly chosen not to use these plugins?

Here we are 6+ months after the release of confluence 4.0 and still no way forward!@#$%

Of course this would not matter a tick if confluence was actually usable without them.

I note with some bitterness and disbelief that Atlassian still promotes this "relationship" with the plugin vendors as a win-win arrangement to the benefit of all parties - not.

My feeling is that Atlassian AND the plugin developers are losing touch with their users at the same time as pricing themselves out of the market.

Can someone please tell me where I have got this wrong?

5 answers

1 accepted

Comments for this post are closed

Community moderators have prevented the ability to post new answers.

Post a new question

6 votes
Answer accepted
JamieA
Rising Star
Rising Star
Rising Stars are recognized for providing high-quality answers to other users. Rising Stars receive a certificate of achievement and are on the path to becoming Community Leaders.
April 15, 2012

It's a good question, worthy of some debate, but probably without pointing the finger at particular vendors.

Installing a plugin is always a risk, and it's going to have some cost associated, I think many users/admins fail to realise this. For instance, it's going to increase the attack surface area, you are reliant on them to continue updating their plugins, it makes admin harder, Atlassian support are unlikely to help if you have a problem caused by a plugin, and so on and so on.

And given that if you have paid no money to the plugin vendor, you have no recourse if anything goes wrong.

But, as you say, out of the box, jira at least is pretty much useless without some set of plugins.

I have some sympathy with the plugin providers - it's very easy to under-estimate the amount of work required to keep a plugin uptodate with all the new versions, coming every month or two as they do, whilst retaining backwards compatibility with old versions. To be honest, I can't blame them for expecting to earn money for all this work, even if their intention at the beginning was to keep it free and open source.

But Atlassian seem quite happy to let plugin vendors provide much needed functionality, so I don't see this situation changing. Take confluence, from an aesthetic perspective it looks horrible. But now there is Zen Foundation, are they going to make any effort to improve it when Zen does it? I doubt it. And if they do they will want to make sure their changes don't clash with Zen's.

The way Atlassian are going with the marketplace, I think the future will be commercial plugins, that you purchase alongside the Atlassian products. These are likely to be supported, and better, because they can afford to prioritise paid development. Atlassian get a cut of the fees so you can expect them to be given a higher profile than free stuff. Free plugins like you have been using will be increasingly marginalised, IMHO, as these are just overhead for Atlassian.

4 votes
Bob Swift OSS (Bob Swift Atlassian Apps)
Rising Star
Rising Star
Rising Stars are recognized for providing high-quality answers to other users. Rising Stars receive a certificate of achievement and are on the path to becoming Community Leaders.
April 16, 2012

As an administrator for over 7 years, I have choosen to take advantage of the many plugins that are available. They have provided great value for our users and company. Many are provided free and some are commercial. Each one needs to be evaluated based on business value, cost, risk factors, and risk tolerance (not much different than any other software acquisition):

  1. What value does it provide? Critical function, nice to have or just a frill.
  2. Does it work? Will it cause administration/support issues with users?
  3. Does it have a track record of improvements and support? Is there a community of users? Is it active?
  4. Who is the provider? Do they have a track record?

Because we have highly customized environment, we are very careful with upgrade planning and testing. We delayed upgrading to Confluence 3.5.13/JIRA 4.4.4 until about 3 months ago because of various software issues with Atlassian, commercial, and open source software compatibility. Even Atlassian has dropped some previously provided plugins/functionality. One open source package needed to be taken over and upgraded - one of the advantages of open source software is the ability/freedom to make it work if necessary. Confluence 4.x will be hurdle (its a major version and UI change!). That includes waiting for all the critical plugins to updated.

As a long time open source plugin developer, it is difficult to keep up with all the various support and improvement request let alone supporting and testing with new Atlassian releases and API changes. Most plugins need to support a variety of Atlassian releases. Confluence 4.x is significantly more work than any previous release. Therefore, many open source and commercial developers are behind where they would like to be! Atlassian did make it possible for many plugins to continue to work in compatibility mode to help with the migration, but that doesn't work for all plugins and it has its own set of user issues. Bottom line is: user community needs to have some level of patience in this area.

Davet April 24, 2012

Thanks for the comments guys. I am really trying to recalibrate the assessments of plugins that Bob so deftly outlines above in order to shield my company from ransomware.

Regardless of whether the plugin is provided free/paid/open source from a large vendor or an individual, if I had foreseen this situation I would have severely limited the extensive list of user installed plugins. Whislt this would have obviously limited the usefulness of the product this would have been offset by the ability to access new bundled features and improvements much sooner.

I believe that Atlassian should issue a warning about the downside of adopting ANY user installed plugin but I know that is not going to happen because in fact the opposite happens. As it is we have some major hacking to do in order to get to 4.1 and at this rate 5.x will be here before we manage it!

I note that the new "Team Calendar" plugin replacing the old Calendar will cost us $2k - extra - p.a. :(

2 votes
Robert Castaneda[ServiceRocket] April 16, 2012

There's a lot of effort involved in keeping up, as noted above by some of the other posters. We worked really hard as well to keep our price points in the "couple of hundred bucks" price range. We hope that we'll continue to see the sales momentum grow as we have for the first 4 months of our graduation launch.

I thought I'd share a few lessons we learned, in the spirit of Share the Knowledge - one of our company values.

Lesson 1: What we learnt is that if you release something and it has your name on it, regardless of if it is free or not - customers will expect you to support and upgrade it within a timely manner.

Lesson 2: Communicate often and promptly - on the day of Confluence 4.0 release, we knew we were going to be up against it as there were HUGE changes. So we communicated this on our community and also setup a survey so that we could collect data on who was using our plugins and more importantly, whether or not they were considered to be "blockers". From our survey results, we found that very few customers care about plugins being open-source vs being supported.

Lesson 3: Build up the team properly - we have a separate support, dev and qa team. There is no point doing products in between consulting gigs - it doesnt work. Quality sucks and SLA's cant be met, the result is passion is lost. Those plugin developers providing plugins in between consulting gigs will feel this pain. They can't plan vacation time, and the code ends up a mess as too many chefs work on the dish.

Lesson 4: Provide an Onramp - regardless of what was written or expectations, customers used our products (even for free!) and we are very grateful to have them. We provided new-for-renewal pricing and then an additional 40% off for those that were early adopters. This resulted in > 70% off.

Hope this helps.

1 vote
Davet October 9, 2012

In the interests of being a good citizen on this forum I have been asked to close this topic and have chosen to do so by answering my own question.

This in no way diminishes the valuable responses recieved but it was always a rhetorical question with subjective answers the only possibility.

i take this opprotunity to close at this auspicious time due to Atlassians decision to approximately double the price of its range which clearly is aimed at inflating an already excessive profit.

As the title of this thread suggests, Confluence and its plugins are inseparable since neither is worth a crumpet without the other.

Adding the price of both together now leads to the inevitable door-opening for competitors so it is off to evaluating that I go.

Greed is such an obnoxious thing, no matter how it is wrapped up in fancy words.

1 vote
Ramiro Pointis
Rising Star
Rising Star
Rising Stars are recognized for providing high-quality answers to other users. Rising Stars receive a certificate of achievement and are on the path to becoming Community Leaders.
April 15, 2012

I can't but agree with Jamie, more than the half of the plugins we are using are free and open source. I can't complain about anything because most of the time the plugins make my day and in cases I'm not paying anything for this. Take the example of Jamie. He's the owner of the Script Runner Plugin and has helped me more than once without receiving much more than thanks and karma points (truly thanks Jamie).

In my point of view, a possible solution would be that Atlassian should pay the developer's copyright, so they can integrate to their products the "most voted" plugins. This would be in order to help the developers that bust their "eyelashes" everyday to help people freely, and in other side to help the customers that burn up their brains every time that there's an update of a product. It is a both side winning.

In case of support you could contact directly to the plugin developer or being redirected by an atlassian supporter. The most important thing would be to have a better connection between the plugins developers and Atlassian.

Just a humble opinion.

Comments for this post are closed

Community moderators have prevented the ability to post new answers.

Post a new question

TAGS
AUG Leaders

Atlassian Community Events