Read this article that Jordan Novet wrote. I know seeing is believing and you will have to wait and see, but I agree with everything Jay Simons says in this interview.
"During the interview, Simons took time to assure me that Atlassian wouldn’t ruin Trello." http://venturebeat.com/2017/01/09/atlassian-is-buying-my-bel...
(disclaimer: I'm the Trello ceo)
And... "Trellists?" They're just people.
I tried to figure out what confluence even does and I'm still not sure.
Considering I'm forced into the atlassian ecosystem at work, this is my plea for y'all to improve.
At that same company, we switched to Jira from 6 or 7 different ticketing systems (a mix of in-house and third party systems split up by department). It was a herculean effort to transition, but the result was much better than any of the individual ticketing systems with the huge benefit of just opening a ticket anywhere and moving it to where it needs to go while preserving the history.
Now the cons I've seen are; I have no idea how much it costs and it seems to require a small to large team to manage and deploy it for maximum usefulness. For ad hoc deploys I've heard/seen performance issues if the server isn't beefy or configured correctly and the UI can be inscrutable if users are just dropped into the default. That's about (the best case scenario of) what I expect for enterprise software, though.
I have used GitHub Enterprise and haven't used BitBucket. I might not be using it correctly, but I don't see what's so great about GitHub Enterprise (especially for the cost). We still have to use a separate ticketing system for non-project related things (same with documentation).
I've used different ticketing systems since leaving the place that used Jira/Confluence, but have been disappointed at what they were using and what else I could find out there. What have you found out there that has worked better?
For ticketing systems, trello or youtrack.
I like Trello, I use it all the time for personal projects, but I just don't see it scaling for large groups. I was surprised to see Unreal Engine using it[1] even if they're only using it to communicate their roadmap (and using something else internally). I thought using Trello was creative and interesting, but I have trouble actually finding and following things.
I use YouTrack at my current job. I guess it works fine for our needs, but it feels clunky and crippled compared to Jira. I'm still using 6.5 (looking to upgrade to 7), am administering it myself, and haven't read too much of the docs, but was surprised it couldn't do some things I expected when I've tried to customize it.
I am mostly used to command line Git during my daily routine, but the designer/manager was happily using Sourcetree. But as far as some products that are kind of worthless, yeah Hipchat would be one. You are way better off using Slack or even WeChat mobile/desktop app (if you are in China)
To my mind, the UX of Github makes it worth the extra expense.
Github UX is weird. It looks nice, and there's lots of nice functionality. But it often takes lots of rummaging to find the things I need, unless I do them every week. Bitbucket and Gitlab seem to be laid out better. Github is the only site I use regularly where I need to refer to notes.
Until recently I would've said there wasn't at all much difference between Github and Bitbucket (aside from Github being the de facto home for open source).
However, I'd now argue that Github is significantly better value than Bitbucket for projects, rather than code. By that I mean, Bitbucket is just a component in the greater Atlasssian ecosystem, so it's not a huge priority to add and improve upon project management tools, as you're encouraged to use JIRA, Confluence etc. Even simple things like Bitbucket's Markdown README's not supporting HTML (in particular anchors for same page links) makes the project organisation experience a whole lot less polished.
Where Github shine these days is that they offer a pretty cohesive experience for an entire project's management. Code reviews have come a long way, as have issues and pull requests in general. More recently they've added support for Projects, which for the most part is a Trello clone with built-in integration into Github's issue tracker. Free hosting of Jekyll websites, previews and diffs for all sorts of non-text file formats etc. really put Github in front. Combine this all with the recent change to bill per user rather than per repo (which provides huge savings for small teams) and I don't at all see how Bitbucket are supposed to compete.
I still use Trello for management of clients' projects simply because my clients don't care about the code. However, I can definitely see myself transitioning entirely to Github Projects and the Github issue tracker in the near future.
Frankly, Trello did really well out of this deal from Atlassian. I know Trello have a lot of users, but I'm a bit doubtful about their ability to generate significant funds (paid stickers...?) With Github Projects maturing Trello were likely to lose a lot of users from the tech industry. Atlassian and Trello actually seem like a perfect fit, so it's a saving grace for Trello who couldn't have timed the acquisition better.
As for ticketing systems, they all suck. All of them. Even Trello. The basic problem is that the amount of information you need to put into a ticket in order to be reasonably useful is more than the users are (time-)comfortable providing. No ticketing software will solve that.
My major problem with Atlassian, though, is the licensing subsystem. It sucks hard. Getting support for it sucks. You can have the account that owns a license, and still be unable to access parts of your own account. They have support for multiple contacts, but only the first one receives relevant mail. The owner of the license can't control the order of the contacts, the contacts themselves have to log in and move themselves to that first position (no potential for abuse there!). Getting support through their support wizard process regarding licensing is painful. Ugh, they need to flog whomever designed that setup... I get the feeling that they used to work for the Windows licensing team....
Their lack of popularity also seems to translate to a lower attack profile, so I feel that my private projects are safer on BB than GH.
I've evaluated many issue trackers too, and none of them seem to come close to Jira. Yes it is slow, but the depth of features (and plugins) are great, and BitBucket works seamlessly with it.
Is there some joke in the middle that's passing right over me?
I guess one way to explain it is that Hipchat feels like an enterprise product, and Slack feels like a consumer product. The difference being that enterprise products get used because some high-level person says, "Lo, all my vassals shall now use Hipchat." Which means that user experience is secondary. Whereas consumer software has to earn each user, meaning that it works harder to please and support those users.
JIRA's BDUF approach to ticketing/bug tracking pleases middle managers whose job is to spend all day clicking around arcane interfaces and finding a way to generate a report that shows their team is highly productive, but it's painful for actual doers to get in there and move stuff around, which means it rarely gets done, which means that the tracking is not very reliable, which means that the value of the bug tracker is dubious. The most important feature any bug tracker can have is that it's low-friction enough that most people will actually use it.
JIRA has tried various things to make this less onerous, including GreenHopper/swimlanes, an attempt to remake JIRA into a Trello-like drag-and-drop interface, but it just never seems to click the same way. For example, today, when I tried to move a ticket in JIRA from the "New" swimlane to the "Done" swimlane, I got a "WORKFLOW EXCEPTION".
Product managers need to learn a simple lesson: if a user wants to use power user features then you should let them because there's a good chance they know what they are doing.
Atlassian abandoned HipChat's bug tracker, and put up a fresh JIRA that they discourage you from using at several steps. That's one way to get the bug count down without fixing any bugs.
Why can't HipChat message editing be as easy as Skype's? If HipChat is at all targeted for devs, why is pasting/sharing code a pita? Where is bold/italics on linux? Your own hyperlinks (instead of pasting raw urls?
I only use HipChat because the alternatives at my company are worse. Even IRC seems better at this point.
No doubt they'll sell someday though so w/e. Temporary / replaceable is the new norm for productivity tools. Beware lock-in! Might end up back on IRC next.
I still remember, and have nightmares about, the day ~300 users lost all private chats, private room membership, and chat history, all because of braindead design.
Hipchat is not my favorite atlassian product. The only reason they have any traction whatsoever is because Slack doesn't offer a behind the firewall product.
The editing function is embarrassingly bad. I get it, I've been on IRC when Perl was cool, but I can't believe they're shipping that. I don't think I've seen a non-programmer use it, ever.
I've also never felt so unwelcome in a bug reporter as in HipChat's. It really reminded me that Atlassian sells to suits, not to users.
just wanted to address my hugest complaint about the Atlassian suite, in the hope someone here can use it for improvement: please unify the administrative UI/UX of JIRA and Confluence (and the other tools). Stuff is named differently, and placed in different positions. Even as a sys-op, it is not always possible to claim access rights to a space that another administrator created. And why does Confluence use JIRA Crowd for LDAP authentication?
For example: https://slack.com/apps/A0F7YS3MZ-jira
1-3 years down the road, it'll be a different story. It'll be unlikely that the executive team will stick around past whatever agreement they signed with Atlassian.
I'm skeptical. I've yet to see this work out with a company I worked at or followed (yes yes I know, Instagram but I never followed them so I wouldn't know how or if they've changed). I'm sure it can and has happened. But you have an uphill battle :)
http://www.businessinsider.com/its-been-1-year-since-faceboo...
Truthfully, unless you are being acquired by an aggregator like berkshire hathaway, I don't get the "nothing will change attitude" you often see. Of course it will change. They bought you to change something (usually about them).
The number that stay truly autonomous is ... very very very low.
And if it goes down a different path, there will be a hole in the market waiting for the next Trello and that's good too.
Good for someone willing to take advantage of the market opportunity, maybe. But is it good for users that have to deal with their company churning from one enterprise to-do list to another?
Familiarity (or lack thereof) with software like Trello can be the difference between enjoying your day to day and finding it endlessly frustrating.
It's not hard to find "______ is being acquired by ________" headlines on HN, where everybody involved promises not to change anything, and then find the corresponding shutdown post on https://ourincrediblejourney.tumblr.com/ sometime a few months later.
Unfortunately, even if Jira were really serious about not changing anything, it's not entirely up to them. No doubt there are some Trello employees who don't want to work for Jira, and a lot of them will eventually leave.
Even if everybody at Trello honestly believes it, in a few months it's not going to be their decision any more.
On the contrary, it went in a good direction under Atlassian ownership.
Exactly. No acquisition ever starts off with, "We're going to change a bunch
of stuff as we assimilate this new company."
I just assumed Trello was too big to be an "acqui-hire", but I really have no idea how big they were.
Fog Creek always pitched Trello as a general list app, not a bug tracker, and refused to add features that would've geared it specifically toward bug tracking. I would venture a guess that Fog Creek kind of accidentally shot FogzBugz in the foot with Trello.
My impression is that Atlassian sees Trello's userbase as a strong opportunity to upsell to the enterprise-style tooling in JIRA and recruit more people into the JIRA ecosystem. If this is indeed the value they see, then it wouldn't make sense to change anything about Trello's fundamentals. They'll probably just create a JIRA plugin and plaster JIRA ads all over the place.
A lot of the time, the executive team that grew a startup into something big are not the best people run that new big company.
Nothing is forever, change is the only constant, cherish the time you had, and so on...
I also feel like (and I realize many disagree) that YouTube largely kept its "feel" and didn't turn into Google Videos 2.0.
Of course there are plenty of examples of exactly what you describe. I feel the same way about Java, Hudson (lol) and plenty of other acquisitions.
Yeah, agreed, I was going to say the same.
YouTube was used as leverage to boost Google+. They backed off soon enough to save things (or perhaps YouTube is just too big to kill with such a mistake) but it was badly messed up IMO.
Plus was the real turning point where I went from loving Google to tolerating them.
It really isn't. VirtualBox was always mediocre compared to VMWare, and ever since the Oracle acquisition, I haven't heard of any major new features or performance improvements coming out of the VirtualBox team. Though, given what Oracle has done to its other acquisitions, benign neglect is a pretty good outcome.
If you want to see the "major new features" implemented, you need look no further than the changelog. [0]
And FWIW, I've tried VMWare Workstation several times and even own a copy of it, but VirtualBox has been and continues to be the most reliable, simple method of virtualization for me (and several of my colleagues). The VMWare drivers/tools are frequently broken by kernel upgrades, the interface is clunkier and more demanding (including requiring hosts to run a couple of background daemons; VBox operates fine on hosts with just the drivers installed), and the performance, while better in some areas, is much worse in others, making it a wash for general use.
Years ago I was talking to a Melbourne dev who said they've made very little (Jira and Confluence I think?) and the rest of it they've just bought and rebranded?
Yahoo likes to kill apps and take devs, like Astrid Tasks.
Would you devote your resources to an Android or iOS app if it isn't backed by a service which is useful in and of itself?
What about desktop Windows apps? Every time I see others interact with a Windows 10 app, I feel like throwing up. Mac apps seem to be doing slightly better, but that might be because it is an inherently tiny market.
Now that everything has moved to the web, people are trying to outdo each other with the creepiest possible tracking analytics. If you wish to develop useful, paid software which is mostly unobtrusive to the users, I would say you are already about 5 years too late.
https://signalvnoise.com/posts/2974-the-slicehost-story
http://www.crn.com/news/cloud/229402851/rackspace-clarifies-...
It always starts with good intentions and fine words about nothing changing and ends with an incredible journey.
Needless to say, Rackspace did NOT get me as a customer after that. I switched to another provider.
Though it's not a profit centre for them, so may be different.
Because the obvious parallel is Hipchat, and while I know Hipchat has a lot of users and even fans still...well.
Being the anti-Jira is what made Trello great. If feels like Luke Skywalker just got bought by the Dark Side while claiming that nothing will change -- I'm still with the rebels.
Your experience will depend on the plugins you've installed, how you've configured custom fields and workflows, your org's SSO policies, the hardware you're deploying it to and how well you keep it up to date (if you're deploying on-prem), etc.
The purpose of software is to specialize hardware to perform a required function. The more configuration is required to do this, the less overall value the software provides for that function.
As a software designer your job is to maximize the sum of that value across all functions that you support. As a software consumer, though, you just want the least possible configuration to perform your required function.
Truth is both tools are just fine. The rest is just people defending their choice as the superior way of doing things.
In other words, like most enterprise software, the experience is highly dependent on how well it's implemented.
Alternately, JIRA as a team board, particularly with remote teams, can be pretty great if you take the time to set it up so that it works for the team (rather than just for the managers or PMs). It can be no more (or at least not much more) cumbersome than Trello, but has a lot of features that Trello lacks.
It has eight zillion features, all of them overlapping. It has so many fields, dimensions, views, and reports that I despair of ever getting a setup that usefully matches an actual workflow. The point of a work tracking system is to bring everybody together, but I regularly see people dealing with Jira by having their own separate tracking methods: docs, emails, spreadsheets, sticky notes on the wall.
I like Trello better because it's more straightforward. Less complexity means less chaos and an easier time getting a workflow representation that matches actual workflow.
But personally, I like things that are more straightforward still. E.g., my last company ran almost entirely on index cards, and we were very happy with that:
Background:
I didn't find a competing product a good match for how I tried to use it, so I pitched that competition (your closest direct competition) on specific improvements (that they should make) but they didn't see my vision. (I was in touch with the CEO and exchanged a few messages back and forth.)
-> Do you have an email address I can forward the same thing to you to?
if you want to after reviewing my forwards you can reply by email. You can also reach out to me at my email in my profile. At the moment I don't use either Trello or them, anymore. (But I do use Atlassian products actually, which as you correctly state in this thread, are orthogonal.)
Thanks - I'd love to use specifically Trello products in the future.
-
EDIT: Why would anyone downvote this? (Let alone 2 people now.) I didn't have trouble getting in touch with the competition... Since mhp is active in this thread I don't see the harm of asking. I could probably find it out myself but I'm not going to waste any more time on this, since it was a waste of time with the other CEO.
The (Startup)ists made (Startup). And the (Startup)ists are going to
keep working on (Startup) at (Acq. Company). (Acq. Company) understands
that (Startup) is unique and beloved and they definitely do not
want to mess that up.
I remain a happy trello user until that happens to your product as well!
Don't be that guy and spin a Jobs-esque reality distortion field.
If things were so perfect to begin with at Trello, you wouldn't be selling. The fact that you did indicates something needed changing.
You hope things won't change for the worse, as do your users, but assuring people that won't happen makes about as much since as guaranteeing you have the winning ticket for next week's lottery.
You can get a better wysiwyg wiki just by sharing a folder in Google Docs and having them link to each other. And I'd take Trello over "greenhopper" any day.
(That of course, or you suck at empathy for your managers, and you're the problem).
The place I work at, Jira is used pretty reasonably. Our devs, in 2 different teams have setup the workflows to show our progress and deployment flow in ways we prefer. The project managers use those workflows happily, they just want them to reflect reality.
I'd rather use Trello, but honestly I'm trying to communicate the work I've done to my proj. managers, so the right tool is basically whatever works for them, not me.
IME, this is most teams. If you get out you are likely moving into another team that's much the same.
Product owneres tried to force developers to use Trello, while we wanted to use Jira with greenhoper.
I really didn't understand what was so fun about Trello, I prefer Jira any time for product planning, Trello for me looks like another sticky notes app.
Pretty nice story if you care to read it.
All these negative comments and hand wringing about Trello being ruined sounds like whining entitlement.
Atlassian is a fabulously successful independently run company that you should be happy bought Trello.
For goodness sake all you haters in this thread make it sound like Trello was acquired by Trump Corporation or The Empire or something.
Even WoW doesn't really do that anymore :)
After that, SourceTree has been slowly starting to suck. It has a bunch of annoying new bugs that don't get fixed, now requires logging into Atlassian to use it on every machine you use, and has had almost no useful or compelling feature additions in recent years. It is basically a good indie app from several years ago, with the Atlassian taint slowly being applied to it.
I dont like Jira either but I do like bitbucket. I started using it back in the day because besides public hosting they also gave me free unlimited private repos and found it does everything I need.
As far as Jira -- the tool itself isn't bad per se, but the way PMs turn it into their own little power-trip kingdom irritates me. There's so much focus on 'control' that I feel like a victim every time I use it. For example, if I want to move a card from In Progress to the Backlog, Jira has an option to restrict the non-anointed from moving cards. So I have to get 'permission' from someone with the right privileges to move a card. Jira is a micro-manager's wet dream.
Trello was the anti-Jira and that's what made it great.
Yeah, this is a pretty common sentiment. Jira is not a great tool, but the things devs really resent is being micromanaged and tracked in such a way that they are turned into a commodity. At the end of the day, if that really bothers you, there are plenty of ways to work free of this feeling- find a company that fits you better or a better manager within your company or strike out on your own. A better ticketing system wont make you happy.
Most things in our JIRA deployment are configured to just allow anyone to change anything, with only a few exceptions.
Locking things down like that is the project manager's way of making sure people are constantly blocked and work isn't getting done. Not sure why some project manager's try so hard to create this state of affairs.
These can be ignored, used to enforce very simple house-keeping things like filling out a "fix version" or "deploy date" when closing an issue as "fixed", or to exactly codify all the rules in a complex development process based on a bunch of flowcharts someone with "project manager" in their credentials spent two weeks painstakingly creating.
[1] What word you choose here will probably depend on if a PM or developers have admin access in JIRA.
For a wiki especially, our team spent ages searching for one that was hosted, private and searchable. Confluence seemed like the only thing with a decent search.
Major drawback of Confluence I've found is that Markdown is not first class, and only supported via some shitty plugin, that renders badly to boot. But the rest of the company (and half my tech team for that matter) doesn't really care about that (the default wysiwyg stuff looks good if you stick to the guard rails).
Still, we're not aware of a better thing we could move to?
If you compare SourceTree to other halfway modern git clients (like Tower or GitX-dev), oh boy does it suck! The layout is completely disorganized, the diffing is super lousy. It's slow. It sucks up lots of memory. It's literally the last GUI client I'd use before just going to the command line.
Yes, I often long back to the days when you could just put software on a floppy disk and be sure that it would never change. That this isn't practically possible anymore is one of the biggest downsides of the cloud.
About the only thing that has had massive longevity is VIM, and at least for me, its throne was taken away by SublimeText. Maybe Atom will replace it or something else, who knows. But in the end, just enjoy the ride while it lasts.
At most, this is welcome news if they manage to integrate Trello with Jira somehow.
(Yes you could buy the Bamboo self-hosted thing, which they're still supporting, but it's clear where their efforts are going at this point).
EDIT: Also, I should add some opinion here. We used to use Bamboo for 2+ years and ditched it (for Jenkins). Bamboo has a horrible UI, and everything basically has to be done through UI.
Jenkins meanwhile has an ugly UI, but you can do everything via config/Jenkinsfile, which is what these newer CI things incl. Pipelines are aiming at also, basically the `ci.yaml` file in your repo idea.
So yeah, avoid Bamboo!
@i386 used to work on Bamboo for Atlassian.
Good luck James, I really want to try out BlueOcean but might depend on stuff[1] at work.
1. GoCD (blah) CircleCI (not used yet but frankly looked better)
But based on what you said, I guess I can try Jenkins over the weekend on a pet project to see how it goes and then compare if it really is too much of a hassle. Thanks!
I have yet to use another ticket system for software development that is as decent to use as JIRA. I'm not saying I love it, but I've done my time with all the usual suspects and I CHOSE JIRA for my company. I found most people who have really hated JIRA for a good reason actually hated the borked workflow that their administrators put them into.
BitBucket has been a great private hosted Git repository for us, we have at least 150+ repositories- the addition of 'Projects' has adequately solved my biggest complaint.
Confluence is fine as far as intranet wikis go, the fact that it's somewhat integrated with JIRA and BitBucket makes it work fine. We tried using Google Pages- never again. Our documentation is easily discoverable and content authoring is easy. We don't try and run anything publicly so our use case is dead simple.
Also, the might include it or leave it separate. They might even include it and still keep it a great product.
My guess is they'll keep Trello running for a while as-is, and slowly rework it to become "JIRA Lite". It will likely have Trello-like functionality with JIRA branding, and some sort of migration path to the heavier JIRA for companies that have grown a lot.
Slack has more speed, better integrations, better search, reliability. (For example, getting notifications on iOS.
I hate HipChat. Apologies to those that love it, but it's likely you never used both side-by-side during normal working days to compare.
So if Atlassian tried to put their mark on Trello, I have little confidence that it will still be Trello.
HipChat is certainly not perfect, but I don't agree about Slack being faster. It definitely has its issues.
Except for when you run into a bug that requires disabling of the Hipchat integration with Confluence... You'd think a company could make its own products work together better.
Atlassian's product range today includes unstructured (Confluence) and structured (JIRA) products.
Trello fits right in the middle. It has a myriad of use-cases[1] and is loved by the millions who use it daily.
[1] - https://trello.com/inspiration
Scott, CEO Atlassian
AWS applies the same trick with some startups in China (like they did with us) and I guess Silicon Valley as well. Bunch of free credits, free for a year. After you have your infrastructure with them it will be hard to switch to somebody else.
If we have built a product that is a (we think) great fit to the Atlassian customer base, who would we speak to in order to create a partnering arrangement with Atlassian? Does Atlassian do such business partnering arrangements?
It's not an addon for an Atlassian product so does not seem to fit to the Atlassian marketplace.
I would have thought that making business connections on HN is just exactly the right thing to be doing.....
Hope to be wrong though!
I am far from expert when it comes to startups, their evaluations and acquisitions, but isn't >1million daily active users [1] not enough to make Trello a billion dollar unicorn? What Trello lacked to reach this? Is it all about the user count and it does not matter they pay or not or you have monetisation strategy or not?
We also used and liked Crucible (their code review tool), but it didn't yet offer some features of Gerrit that were very important to us, so we moved to Gerrit.
Luckily there are free software alternatives to trello you can self host that are not going to sell out such as wekan[1], taskboard[2], taiga[3], kanboard[4], ...
[1]: https://wekan.io/
[2]: https://taskboard.matthewross.me/
[3]: http://taiga.pm/
I like Jira but I understand why others aren't fond of it. I'd argue it completely depends on how you configure/use Jira. Administering it is annoying regardless though.