What I find shocking about this, as with many other examples from Meta, Alphabet, Amazon etc. is that many of the complaints are things that small companies justify as "yes, but small team". In other words, our 3 developers don't have the time to refactor this API heavy interface because we are struggling with company growth.
When you have the size and money of Atlassian, that excuse is pathetic. Yeah we have 3000 Developers (or whatever) and we don't have the resources to do good UI design, refactoring away from poor performance etc. I always thought that these companies were the ones who invented "how to do good web apps", but it seems like function over everything and even the functionality is poorly implemented in some cases.
How can anyone think that a simple app that downloads 24MB over 900 connections is OK. What the heck?
I think you're hitting on something there. I think most of those big companies are struggling to provide meaningful changes because they are too big and there is no trust. Trust and actually knowing your sub-ordinates is replaced by tools and metrics. Those tools and metrics then drive exactly the opposite of the behaviour that you would want.
Large companies incentivise behaviour that overall make products worse and customers angry while showing super awesome metrics internally. While I'm sure such metrics are going to be causing some of the right people to get PIP'd and managed out they also disillusion the good people that were trying to do the right thing (or actually did do the right thing) but which didn't show up as the correct number on some dashboard that was used by the CTO and HR to fire the "underperformers".
> I always thought that these companies were the ones who invented "how to do good web apps"
No. These companies were the first to get a critical mass in their respective markets or product categories. then they used that position to smother or buy out competitors and ensure their survival.
There is nothing mythical or extraordinary about what those companies did, just some mediocre people who were in the right place at the right time.
I am a consultant and have tried to use Jira as a tool for managing the work of my team. I find if you have more than 10-15 tasks, I just loose the visibility. It’s easier just to create a simple Word document with structured bullet points (first indentation level is the topic, second level is the outcome, third level is the task). One weekly plan, one A4 in Word. When a task is done, I cross it out. When I need to prioritise, I tag each task as p1, p2, p3. It works SO MUCH better because of its simplicity and flexibility. In addition, I can also easily convey to the managing partner / VP what the team is working on without needing to click around 15 different cards and sub cards.
I’ve used a word document for team organization before and I think the depressing thing is that as inefficient as it is, it’s not really all that much less efficient than anything else. The more structured and fancy something is like Jira, the more it insists upon itself.
I swear Confluence was designed by somebody who woke up from a coma or cryogenic sleep and brought cutting edge 1996 functions to the web. It was so sad to see people like it. Oh you poor pitiful souls…
Bitbucket server was actually getting good when they cancelled it. They upped the prices of data center to dissuade people.
Their Bitbucket cloud is a travesty, once a promising place, offering private repos before GitHub... But they squandered their lead by letting it languish.
I would be happy if Jira died in a fire, but the fact is that if that happened, middle management would simply find some other software to use to micro-track their pseudo-agile detailed upfront plan.
The issue isn't Jira; it's the question: Does Jira work for you, or do you work for Jira?
You can judge the quality of a project management tool based on one simple metric - when you sit down to use the thing, do you spend your time thinking about the actual content of the work items? Or do you waste time in the meta-structure of the work, effectively shuffling tickets and story points like it was a game?
Jira provides endless ways to engage in useless meta-work, it's nearly the entire purpose of the UI. Style over substance is the result. It feels intentional on Jira's part to woo their target audience - mediocre managers who struggle to produce well-written and well-designed technical content. Dress any plan up with enough fluff and it looks impressive.
> It generates amazing reports, fantastic predictions, fancy charts, great managerial tooling, makes imposing processes easier, has robust integrations, etc., and that’s why most managers I’ve spoken to really enjoy Jira, while on-the-floor engineers either range from fiery-eyed hate to toleration.
You kinda need to impose process for large projects, and know where you are. The engineers are free to show a better tool for the job.
> I think Jira does a good job of making teams look much busier than they are, but in the end I see it more as a “talk more not do more” situation.
It does not make teams looks busier, it shows that teams are a lot busier in reality, which is valuable information.
> The Alternative
> Although a potential downside to Linear is that it is built explicitly for engineers, thus may not be suitable for other teams such as marketing, design, etc.
And this is why engineers will never build any viable alternative to Jira, because there is this us vs. them mentality. Issue management tools need to work for the whole organisation, that's why Jira is still the gold standard.
> If you’ve given Linear a try, but it just doesn’t work for you, some viable alternatives include GitHub Issues, Asana, ClickUp, YouTrack, Trello, Pivotal Tracker, and monday.com.
None of these are viable in larger organisations, they are fine for small projects, but you would have a lot of trouble with them in an enterprise context.
Jira's web UI can be very annoying, however an alternative to Jira would need to be at least as customisable and familiar as Jira, otherwise no large team will ever adopt it.
Author here, for the record, I never said the problem was the imposition of processes, I know that is totally necessary, and was listing great "pluses" to Jira as a management tool.
Something I think I could have iterated better in the blog post was that I believe Jira--even without all the bureaucratic processes it promotes--panders solely to higher-level management with zero regard for the ones interfacing with it: the engineers. That's the big issue I see.
I think Linear--by stripping away all the hullabaloo--continues to provide management good insight on team progress without abandoning the ones who actually use it on a daily basis.
I'm really hoping Linear continues to step up to the plate and manages to take on some of the features that make Jira a more attractive piece of software for a lot of people that are familiar with it veering Linear away from its great UI, speed, etc. However, I wonder how many people are actually picking Jira based on its merits rather than it just being what they're accustomed/used to, or simply not knowing there are other options out there.
I feel that the main issue with Jira is that it is a victim of its own success the same way Windows was, when it was built it fulfilled the needs of the time, however time did not stand still and instead of refining what they had, they kept on adding to it without changing its fundamentals.
I think we both agree that in the age of VSCode: it is unacceptable for 30+ billion dollar company to have such a sluggish web UI.
I'm an engineer and I actually feel that Jira fails me because it's not flexible enough, some examples include:
- issues cannot have more than one assignee
- projects cannot have multiple releasable items / components
- poor Bitbucket integration
> However, I wonder how many people are actually picking Jira based on its merits rather than it just being what they're accustomed/used to, or simply not knowing there are other options out there.
I pick it still because most people will understand how it works and I am able to customise it the most to fit everyone's needs more than all the other tools out there.
I have been promoted to Team Lead recently and while I saw Jira as a burden while being a dev, I now totally rely on it to keep on overview.
We're currently looking into outphasing Jira and going to GitLab, but as I see, you can't nest tickets there. Can you? This is absolutely crucial to me. I'm happy for recommendations that allow ticket nesting (Epic, Issue, Subissue) and interface nicely with GitLab.
- Literally never been a problem for me, but maybe not everyone is always connected to gigabit ethernet
- Just go to the issues browser where you can see and search for literally all the issues?
1. Jira speed is rarely limited by internet bandwidth. It’s a very slow product in many implementations. Your gigabit internet likely has nothing to do with the performance you experienced, but also correct: not everyone is on a gigabit connection. But if you need a gigabit connection to experience Jira properly, that’s unacceptable.
2. Is your experience with a vanilla instance and/or a small ticket volume?
The other thing that seems to commonly happen is the customize-ability of these systems leads to major per-transaction overhead/bloat as custom business logic on its 400th revision is executing with ever more db-impacting logic.
- You can now trigger Status to Done
after merging Github pull request.
- If someone assigns you a story as
reviewer, you can trigger the Status to
In Progress (if not already).
- You can have a story automatically
reassign to the reporter once it's
marked Done.
These scenarios reduce the need to switch tabs or click, click dropdowns, which is nice.
It would be nice if we could do requirements tracing from business all the way to ITSM change request, track associated defects--some combination of semantic analysis and code grepping--and ultimately dashboard all this.
It seems like it would approach the model of a thinking organism.
Author is part of the problem. If his job of status'ing the stuffing
out of everything didn't exist (and it largely didn't until 2010),
there'd be no Jira and a lot more front-line workers.
There are various command line clients and you can build your own tools with the REST API.
I wrote a simple command line script that outputs my time tracking on JIRA issues because our time tracking tool (that read the data from JIRA) is even worse than JIRA itself.
I get that some ui issues are hard to solve given that Jira is heavily customizable and backwards compatibility is a real issue. But for ui performance there is really no good excuse.
Sheesh, is this really what we're voting to the top of HN?
Don't get me wrong, I can be as frustrated by Jira as the next person, but this is simply a rant fuelled by the writer's anger at internal processes as much as the software itself.
Additionally, the numbers on data usage and network requests do not align with what I see.
It's just a couple of fair points surrounded by populist "yucky Jira" fluff.
It's like people conplaining about SAP. SAP also is hard to implement, because it's hard to map "loosy" business processes to a perfectly integrated and managed system.
Yes, Jira can be pain in the ass. Because you are not able to adapt your workflow. Jira offers a thoughtful solution to manage your progress and processes.
I use Jira and agree with some things on the article, but for other client I have to use Azure DevOps and please give me Jira.
Also other thing from this articles "I hate [tool]" is that many people just don't like their job or are burned out, the hate may be coming also from that, not only for a poorly designed tool.
Author here, was just Googling my post to figure out why it was suddenly getting tons of views. FWIW I love my job, not burned out, but yeah... dealing with Jira has been a serious pain in the butt, especially coming from Linear.
My pet peeve with Jira is workflows that do not reflect the actual code delivery workflow. It leads to every ticket constantly being in the wrong column and makes the stakeholders observe and interact with an imaginary abstraction of the delivery process.
I find jira great, but we use it to track a task by adding comments to it, sometimes assigning it to ourselves, and then eventually closing it. Nonsense such as states and workflows and resolutions and priorities are meaningless distractions.
I agree that for bugs a simple list may be the most valuable or at least offer a good overview, but bugs are a world different from what is needed for feature planning / epics / sprints and so on.
Aside from Trello and friends, Azure boards is actually also pretty OK, if you disregard their very confusing naming and weird menus. It's fast and easy to use in the board view.
Everything Atlasian sucks so bad. You can force the products to get the job done at times, but the customer service and support is abysmal. Insulting to put it mildly.
Github loads fast and is usable. It might have less/other features, but why would having more filtering options and more fields in each "issue" make load times 30x slower?
Same with confluence, I don't mind if the editor takes a minute to start up, but why reading the page takes like 10sec?
JIRA sucks indeed, but I have seen worse. My pet peeve with Atlassian is that they purchased Trello, and I was really rooting for Trello to refresh the landscape....
There are none, because Jira does a million different things, not necessarily all that well though.
Jira can be a task manager, a bug tracker, a service desk ticketing system, a project planner (regardless of methodology applied), a Kanban board, a knowledge database and so much more if you know how to configure it. That's also why sucks in many cases. You end up in these weird situations where you have a service desk ticket marked as an Epic, or on person on the team uses a kanban board, while the rest are just going by a different priority. Sometimes priorities migrate from you service desk to a SCRUM project and things make no sense.
You can work pretty well with Jira, if you can stick to one set of feature and keep the entire organisation locked to one type of projects. There are some things it's not good for, I don't see it as being a good project planing tool, regardless of how you try to use it. As a service desk tool it's pretty good.
I will say Jira is much better something like Phabricator or Bugzilla. Neither of those are alternatives, Phabricator is the closet, but it still doesn't do all the things Jira does.
Linear is gorgeous and satisfying to use, very aesthetically pleasing.
It includes Jiras killer feature, links between issues, across projects.
But...
It can't sync issues with github, it must fully be the source of truth for everything. This made it a non starter for us, we tried but it just can't work when not every piece of work by everyone is in linear. Even Jira allows sync.
It also didn't allow enough detail. For lightweight task management it was great, but when you have big discussions it was weak.
Github is great, but is missing issue links and better lightweight planning like tasks lists that auto update when linked issues update.
It’s been several years since I’ve used Jira so can’t give an honest head to head comparison, but my experience with Linear has been fantastic. Looks nice, has all the functionality I look for, enough customizability.
If you’re looking for something you can configure every little bit of for a very large team (like Jira) it may not meet your needs, but otherwise I highly recommend.
