Tillywork: Open-Source Work Management | Hacker News Reader

Tillywork: Open-Source Work Management

github.com

118 pointsthunderbong33 comments

Neither the website, nor the github page show you what this actually does.

> Manage timelines, allocate resources, and monitor deliverables for your projects.

Yes but how? Some screenshots or a video might help. Even some more detailed text.

I'm not going to compile every random project just to see a "quick" overview of how it actually works.

I have noticed this a lot with open source projects, for some reason they seem to be against any amount of meaningful screenshots in the documentation.

I think we can accept that the screenshots may at some point be slightly out of date and won't always be updated for an open source project unless there is a major change.

But we can't treat a GUI based application like we do a CLI one. They also don't have to be super professional screenshots like I see on some projects. Just some simple screenshots so I can at a glance see if it is something I care to try to download and setup.

The github links to https://tilly.work at the very top right - not below the fold. Screenshot there.

The image wasn't there when I posted my comment.

I think it got HN-ed to death

This looks promising. I'm occasionally interested in trying out different project management software. The challenge lies in overcoming the momentum of the specific tool we're already entrenched in at work. Even if I discover a superior tool, it's unlikely I'll have the opportunity to implement it. Has anyone managed to adopt a different tool successfully, diverging from what the team is using, or is it simply unfeasible without team-wide adoption?

My current workplace is pretty lean and flexible when it comes to this kind of thing, and we've tried a few different project management solutions over the years. Without fail, the new solution looks great at first glance and feels great as we start using it. Then it gradually becomes annoying and horrible just like the last one, and people start yearning for Jira. The horrible-ness usually comes from straight up bugs, a lack of comprehensive reporting, weak integrations with other tools, and/or a UX that doesn't jive with certain roles/people.

I think it'd be a real challenge to meaningfully compete with Jira. You'd need to not only match its feature set but also be faster than it and less buggy. It seems that being a "not Jira" gets you a good amount of customers off the bat due to how much people loath Jira, but it'll probably take a paradigm shift or a very competent team to overthrow it.

IMO Jira is the epitome of a tool being what you make of it. If you keep things simple and pick the slice of features that work for you, don't go crazy with plugins or creating extra ticket fields or mandatory workflow restrictions, you'll be fine.

If you have multiple PMs each trying to impose their own view of the world expressed in ticket form, it'll be a fiasco.

Very true. I am a reluctant user of JIRA, and once I was past the realization that JIRA isn't meant to be used out of the box, but rather meet your process and production where it is, it kind of fell into place, and there are few pieces of software that can remain straightforward in one area, and be good at absorbing details and complexity as things pop.

Linear looks pretty interesting too.

I'm curious, what reporting is actually useful to your team.

I used to work on a project management tool, and everyone wanted "reporting", but when you got down to what they needed it was usually pretty minimal.

The "Will it be done before/at the unreasonable deadline?" report.

Indeed. I think the most interesting and worthwhile reporting hinges on having pretty rich metadata attached to the tickets, so you can answer questions like how many man-hours did feature X take? how much over the estimate was that? how consistent is that compared with prior similar features? stacking up the critical path from now until release, and given the present estimates and past history, what timeframe are we looking at?

But to really get anywhere with that you need everyone to have put in the required metadata, to be correctly linking tickets that block each other, to be providing estimates upfront and tracking time, to be accurately capturing the work in tickets. You also need people not to be sandbagging and then using the "extra" time to perform localized refactors and other cleanup that they can't otherwise get buy-in for.

This was exactly what LiquidPlanner did (probably does?). I can't speak for it today, but it was built as planning and scheduling first, task management second.

As a bonus we handled everything a ranges (that task? 1-4 days) so you'd get results like: this project is 90% likely to be done by a given date, 98% by this much later date you're going to hate.

$15 monthly per user, and I can't even demo the software without scheduling an appointment? LiquidPlanner might be the greatest software in the world but I can not schedule an appointment for a demo - my opportunities are far too spontaneous for that. This seems like software for people who like to schedule things - I need software for people who are forced to schedule things.

I don't actually have this, but the simplest thing (in my head) isn't possible natively in any tool I've used: I just want to report on "we did what we said we would do". That is: - What did our team agree to do in the sprint (added to sprint at planning)? - What got deprioritized (moved to backlog)? - What got introduced during the sprint (added to sprint mid-sprint)? - What got completed? - What rolled over into the next sprint?

Obviously, I can audit every single story that was modified during the sprint, and/or make spreadsheets or custom BI reports, and stuff... but I don't wanna!

Right. It does feel that way. Jira is the worst project management software, except for all the others.

Perhaps instead of new tools to compete with the entrenched, we should be building new frontends.

How many people forced to use JIRA or Salesforce would jump at the opportunity to have a better interface?

(I work on a tool completing with the entrenched)

Building one’s business solely dependent on the grace of another’s API access is a pretty precarious position. Beyond that the incumbent APIs are usually slow and severely rate limited. Building improvements on top would necessitate a complex sync/caching system, plus if you want to layer on real time collaboration your job is even harder since you need to maintain your own state (for collab in your CRDT format or whatever) that you periodically reconcile with the underlying tool. At that point, you already need a complete backend so you might as well be a stand alone product and just offer 2-way sync with the incumbent as a value add feature.

Anyways, we built something like that https://www.notion.so/help/guides/synced-databases-bridge-di...

A different frontend wouldn't fix Jira. It's slow because the underlying APIs are slow. The frustrations mount mostly from the sheer number of rules that are built on top of it by users; if your alternate UI obeys those rules it will still frustrate users.

Have you found any you’d want to swap to?

For open source project management tools I prefer to use the integrated solution from Phabricator[1], nowadays called Phorge[2]. It has workboards, very good access control and separation of projects (tags). See an example board[3].

1. https://www.phacility.com/phabricator/ 2. https://we.phorge.it/ 3. https://we.phorge.it/tag/phorge/

I see lots of people sing its praises, but (from the outside) it always seemed like it was kind of in limbo, not exactly abandoned but not exactly going somewhere either. I do see some recent activity <https://we.phorge.it/source/phorge/history/master/> so it's being developed by someone but I think https://we.phorge.it/T15801 squares with my concerns - there is likely a marketing problem for folks who aren't already in the know

CRM is a bit of a weird pivot for this, kind of lands you in the same space as Productive.io, which at face value is a really weird combination of features .. but actually is in a kind of ideal spot in the current market for small agencies and consultants.

IMO would have made sense to build out CRM before task tracking though.

Plane looks much better, despite being a ripoff of Linear

https://github.com/makeplane/plane

thanks for the mention. it's not great to hear we are a rip-off of Linear, but i get why you are saying that. there are quite a few differences between Linear and Plane starting with how unopinionated Plane is compared to Linear, but we need to do a better UX job to call that out in the product.

work's ongoing. expect changes soon.

An open source option isn't a bad thing.

Is there something that this project does that the opensource gitlab doesn't?

nice!

I work on NGO projects with a lot of multi-lingual volunteers, and so far have been struggling to find a good and simple OSS tool to work with it.

Really like the simplicity.

2 things that would be good to have: - SSO (I am using keycloak for a few clients - it's simple and allow user not to get confused with ten different login one for each tools we inflict on them) - tags with filter by tags (I see the filter feature is coming so I guess you might be working on it.)

FWIW, I don't see yet how the CRM would be useful but maybe for customer support and follow up - could be nice.

Congrats anyway!

surprised you haven't yet heard of Plane---we need to do something special for NGOs, :-)---, but do give us a try. https://git.new/project

what you are looking for and more's already available.

hey, thanks, I will have a look.

I actually can't get past using Excel as a Gantt sheet

need more tables like the duration for a task to complete. and vikunja looks far more promising than this IMO

Coming after Salesforce, a bold move!

Reply on news.ycombinator.com