Hi @jrymer, thanks for jumping in! Looking at the demo again, it’s not as slow as my first impression now that it has been cached in my browser, but it still “feels” slow to me. I think part of it is animation choices, part of it is the lack of loading indicators / loading states, and part of it is the lack of optimistic updates. I think all the blurred box shadows also look a bit dated.
Though if it helps, here is a video showing it takes about 2 full seconds to load and display an empty project; it “feels” a bit slow to me but I guess it’s not that bad really. The apps we’ve been building with React + React Query do feel a lot snappier though.
I understand your point, but Jira at the end of the day is proprietary, we will have to compromise somewhere and that might be in clunkiness. In some or the other way JIRA will always feel like an upgrade going from most of these software because it’s the most “full-featured” of them.
I will give worklenz a closer look, and also look into odoo. I looked into Leantime and it is an open core product as well. I’m not entirely sure about odoo yet as it might be open core. It’s an ERP/CRM software that also has project management so it might potentially consolidate more of our tools if it works out.
I looked at huly, and it seemed nice, but a bit too focused on AI. Still, it was on my list for a deeper look as well.
I’ve had a briefish look at odoo. For me, odoo was attractive since it is written in Python, so we already have the expertise to work on it, fix bugs, and contribute features. I checked it out and it seems pretty flexible, and will support things like:
Managing time off
Managing invoices
Managing quotations
Managing leads
It is an open core product though, and they have enterprise modules and features that are not available in the open source version. They have a marketplace of modules of all kinds, open, free, proprietary, paid etc. They don’t allow posting modules to their marketplace that compete with the enterprise ones, but I think it should be possible to sell/share them off their marketplace.
While it is open core, it also has the capability of consolidating more of our process into a single tool. The open source bits have enough to fulfil quite a few of our requirements. It has proprietary modules; however, there is also a non-profit org associated with it that is focussed on the open source core product.
I’ve had a look at Huly and I’ve come away mostly impressed.
Huly can handle most of the stuff we user JIRA for and go a bit further. To start with it has the concept of workspaces which are like isolated containers of all Huly features, it’s closer to multiple instances/tenants, but for our use case we could use it as an alternative to our cell projects and have a separate workspace for Bebop, Serenity etc.
Huly does not have epics, but if we use workspaces for teams we can then use projects with unique keys instead of epics. Or there is a similar feature called components that we can use.
Huly does not have point estimates, but we can add custom fields, it’s pretty flexible. It does support hour estimates and time logging. It doesn’t have a reviewer field, but again we can add custom fields for that. It has the concept of “collaborators” which allows adding multiple people that can be used as reviewers, but we can also just add two reviewer fields or single reviewer field that is an Array of type Person, which is something it allows you to do and we can have an arbitrary number of reviewers. Huly also has customisable states for tickets, so we can add In review, Blocked, etc.
The biggest things Huly does NOT have is sprints. There is the concept of “Milestones” but those are per-project and only have a target date, not a start date.
There are some features that Huly has that JIRA doesn’t (not the version we are using).
Huly has a “to-do” list of sorts that is calls “action items” and it automatically creates one for each task that you move to in progress, you can then schedule it in your planner. You can also add lightweight todo items like “reply to email” or “try xyz to debug abc”.
It has a calendar for marking vacations availability etc that can sync with google.
It has a confluence-like document editor.
It has a Google Drive like space for files.
It has a mattermost-like chat space for discussing tasks.
It has some CRM-like features that allow managing companies, contacts, leads etc.
It can sync to GitHub so we can have all out github repo issues show up as tasks automatically.
Honestly, other than the lack of sprints, it’s a pretty good impression I got.
However, what I find a bit off-putting is their upcoming pivot to the blockchain. They want to make either a new platform or an evolution of Huly called “Global Huly” that is fully on the Blockchain. This to me is a severe red-flag.
Wow, parts of that read like an April Fool’s post. I do think the parts in “Breaking Down Organizational Barriers” are very insightful - it would be awesome if we had more flexible ways to collaborate with clients, Axim, etc. on issues in our issue tracker (and theirs). Slack and GitHub allow this but Jira does not. However, the blockchain stuff seems ridiculous.
This has been a pretty hard discovery for me to do. For some reason, I thought this would be a solved problem in open-source, however, I don’t see any perfect match that we can adopt without compromise and an expensive migration path.
No matter what path we choose it will take a lot of time and effort to move over, a lot of change in processes, change in scripts etc. However, ideally now these changes would be part of the tool we use instead of a patchwork of services and scripts.
@team I’ve created a ticket for everyone to review this discovery. Feel free to pull in into the this coming sprint on the next. Once everyone has given their feedback here or on the doucument, we can start investigating the replacement and estimating the effort involved in moving to it.
@kshitij Sorry to be late with this, but I do think there’s one more that’s very promising: AppFlowy is an open source (AGPL) competitor of Notion. It supports self hosting, has a very active open source community / development, has a sprint template, and could also replace our horrible CRM. It doesn’t yet have integrated time tracking but it does support time fields on issues (like time budget, time spent, etc.) and I suspect it would be straightforward to build time tracking as a plugin.
Whichever tool we pick, it might be good to start with a PR to the project and see how it gets handled?
Hello again @braden ,
Thank you very much for your feedback and the video that you made, it was a huge help into pinpointing the heart of the matter.
We were testing for huge data load but always using an awesome connexion with our servers close by hence we never noticed those huge loading times. Constraining our network traffic speed is now definitely part of our Q&A process!
We have made a new release last week, starting with performance enhancement (here is the announcement), particularly regarding the initial loading time that your video has pinpointed.
It is still very much a work in progress (the project is but a few months old after all) but I think you’ll find it much less clunky now if you have another try at it!
And sorry again to have boarded this topic as a stowaway haha, I was pretty mind blown to see Tenzu mentionned at all.
There are a lot of these more free-form build your own data-structure kind of tools like Grist, NocoDB, BaseRow, Twenty and possibly many more. I avoided them because I had tried them for an unrelated reason a while back and while they are very flexible, it can be hard to get them to work the way you want. Though perhaps they are all worth a look?
I can only speak for myself but I’m very glad you posted here, so no apologies needed! Our forums, and this discussion in particular are public exactly for this reason
In fact that just reminded me that the document I posted was supposed to be public as well, so I shall do that ASAP.
I think that can be quite time-consuming, so I don’t think we will have the budget for everyone to do this individually. I’ve allocated about 30 mins for going through the documents and hopefully that will help you figure out which ones are worth a look at. I plan to create sandbox servers for these at a later stage.
Hi @jrymer, nice work! I can definitely see an improvement already Thanks again for stopping by and for jumping in to the discussion - we really appreciate it! Don’t hesitate if you have anything more to add to the conversation.
I have most of these tools installed locally, so I’m thinking of making short videos that show the UI and workings of these tools.
I don’t want to spend a lot of time, so these will be entirely unscripted and without narration. I’ll just perform some basic tasks so we can get an idea of the flow.
hence we never noticed those huge loading times. Constraining our network traffic speed is now definitely part of our Q&A process!
Thanks again for stopping by and for jumping in to the discussion - we really appreciate it! Don’t hesitate if you have anything more to add to the conversation.