Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Untitled topic

news.ycombinator.com

Post reply on HN
Print view View original

51–60 of 136 posts

Re: undefined

#51

Post by glodalitazic »

SOLAR_FIELDS wrote: post #29
gwervc wrote: post #20

> Then we use AI to automatically create a title from the description, so you don't have to spend time figuring out a title Yes please, tickets that POs spent even less time thinking and writing about.

While I can detect that perhaps you don’t feel like this is a useful feature, this is absolutely a legitimate and useful feature. I use it literally every day with Linear. With Linear you can make a ticket in Slack by clicking the … and selecting “make linear ticket”, which gens out a ticket that summarizes the title using AI and pulls in the conversation that you genned it out from, including future updates. The ai…

Imo thinking about the title actually helps you to get a clearer view of the problem

Re: undefined

#52

Post by cess11 »

Jira is extremely customisable. The "too complicated and enables a lot of processes and micro-management which used to kill our productivity" thing is a matter of management in the company, or whoever controls the implementation of project management flows, rather than Jira itself.

I'm not a fan of Jira and avoid it if I can, but it has never reduced my overall "productivity", putting some text in some forms and clicking some buttons just isn't the kind of factor with such an influence in my experience.

Re: undefined

#53

Post by ffsm8 »

> We load all the data from local (indexed db),

Ymmv with that. I've written a webnovel reader for myself before (never published it because of the obvious copyright issue of scraping the content)

My first approach was a Django SSR app, and it worked great for years. Since then I've remade it several times in various languages, attempting to get the smallest latency possible vs the developer experience.

In my testing, web requests were ultimately faster then indexeddb if the query was essentially pre-computed / simple select (i.e. persisting count attributes with via insert trigger etc). My currently deployed version is java, and it's only role is to extract the userId from the jwt and feed the request params into the query, which is serialized to json by the DB, so the java application simply returns the data as string while setting the application/json media type.

I got full page load (pwa, so cached html/Js) to 8-15ms with that approach, querying the data from indexeddb took upwards from 25ms on all of my attempts, usually 50+ms as the amount of data increased later on

Re: undefined

#54

Post by yaswanth »

AI for project management is an amazing use case. May I suggest adding features:

1. Expanding on the description taking the project meta info or history into account. 2. Adding to the above, come up with feature scenarios and test cases 3. An agent to listen into standup conversations to come up with an updated priority list for PO to approve? 4. Move tickets or suggestions to move tickets to next stage of workflows depending on the state of the ticket? 5. Optimising workflows and suggesting better workflows listening into stand ups, retros etc. 6. Come up with demo list of tickets for team to demo 7. Prefill or suggest topics for retros 8. Come with timeline suggestions depending on history and evaluating the code and designs 9. More on code, come up with architecture diagrams (puml etc) and API contracts dependant on code, tickets and conversations 10. Come up with sub tasks/suggestions after evaluating the description and code base.

Re: undefined

#55

Post by mrweasel »

cess11 wrote: post #52

Jira is extremely customisable. The "too complicated and enables a lot of processes and micro-management which used to kill our productivity" thing is a matter of management in the company, or whoever controls the implementation of project management flows, rather than Jira itself. I'm not a fan of Jira and avoid it if I can, but it has never reduced my overall "productivity", putting some text in some forms and clic…

It's the same issue every time someone tries to brand themselves as an alternative to Jira. They implement 10/20% of the features, typically part of the task/bug-tracker and maybe the KANBAN boards and completely ignores the rest.

The fact that products and startups feel the need to brand themselves as "Jira alternatives" says a lot about the absolute dominance of Jira in some markets. You're just not going to replace Jira with a minimalist bug tracker. If profit is your goal, you'd probably be better of reimplementing your AI feature as a Jira plugin.

Re: undefined

#56

Post by zerop »

Good stuff, like the AI approach to this.

On Jira, I keep hearing the criticism of Jira on this forum a lot, but IMO it's not that bad product. Jira is used by all industries, not just by software management. Making a tool has to be a "silver bullet" for managing any kind of project is not easy. Tell me a better PM tool which is as widely used as Jira.

Re: undefined

#57

Post by fidotron »

There's a ludicrous aspect to software development, which is if you ask around about early uses of computers you'll run into lots of people doing critical path analysis for other industries, yet we have persisted in pretending software development is somehow unique and does not actually work like this.

The whole idea humans should be prioritizing tasks has been wrong from the start, and indicates a lack of the right information in the system.

Re: undefined

#58

Post by KronisLV »

steve1977 wrote: post #37
harshithmul wrote: post #18

Earlier quoted context omitted.

Indeed, our user interface (UI) shares similarities with Linear’s. We, too, are fans of Linear and appreciate its user experience (UX). After speaking with a lot of people, we concluded that it’s top-notch. Consequently, we wanted to give that experience rather than attempting to reinvent the wheel. Despite drawing inspiration from Linear, we built everything from scratch. Our primary focus is the AI and task priorit…

> Despite drawing inspiration from Linear, we built everything from scratch. If I was to reimplement Excel from scratch but make it look like Excel, Microsoft would still sue the shit out of me.

LibreOffice exists, doesn't it? In particular, Calc: https://en.wikipedia.org/wiki/LibreOffice_Calc

WPS Office is even more similar in some ways: https://www.wps.com/office/spreadsheet/

FreeOffice is also pretty much like Excel: https://www.freeoffice.com/en/features/freeoffice-planmaker

OfficeSuite feels like a carbon copy: https://officesuite.com/en/sheets

There's also a variety of online offerings, yet someone every single one of them isn't sued into oblivion.

Note: I can personally only endorse LibreOffice, but either way, there's many similar products and that's not necessarily a bad thing either.

Re: undefined

#59

Post by harshithmul »

yaswanth wrote: post #54

AI for project management is an amazing use case. May I suggest adding features: 1. Expanding on the description taking the project meta info or history into account. 2. Adding to the above, come up with feature scenarios and test cases 3. An agent to listen into standup conversations to come up with an updated priority list for PO to approve? 4. Move tickets or suggestions to move tickets to next stage of workflows…

thanks a lot. We will add these to our idea board, definitely a lot more work to do and things to build

Re: undefined

#60

Post by harshithmul »

ffsm8 wrote: post #53

> We load all the data from local (indexed db), Ymmv with that. I've written a webnovel reader for myself before (never published it because of the obvious copyright issue of scraping the content) My first approach was a Django SSR app, and it worked great for years. Since then I've remade it several times in various languages, attempting to get the smallest latency possible vs the developer experience. In my testing…

We also have tried initially to build an SSR and get data from the server but when in the process we figured every time making an API call was slowing things down and adding up a lot of latency even if start adding complicated algorithms to the API call caching.

We think indexeddb also evolved in the recent years, but we will definitely keep an eye and keep improving with the available technologies.

Post reply on HN