I think higher management likes to believe that those things (sprints, scrum, Jira and standups) provide a safety net against lazy employees not working hard enough, so they cling to them. Of course, they actually do little and are pretty easy to game. Their failure to magically make all software work predictable and deterministic and all developer time fungible actually means that you still need a manager involved and close to the work to identify people who BS their way through everything and take way longer than they should.
I'd rather be in an environment where you are given access to simple tools like kanban to prioritize and track work, and for people who don't deliver, the manager just fires them (maybe with one warning).
I've been in the software business since 2007, which was also when I first met Jira and Scrum (at least something with 14 days sprints).
My first encounter with Scrum (or whatever it was) was good. It felt good to work in cycles and reprioritize twice a month.
Since then I have seen various versions of working systems and various versions of broken systems.
The two last projects have been extremely agile, the current project has exactly 5 mandatory meetings in an average week:
- 3 x stand ups that typically take <10 minutes and never more than 15.
- 1 stand up plus planning (scheduled 1 hour, typically takes 20 minutes)
- 1 stand up plus voluntary demo + retro (scheduled 1 hour, typically takes 30 minutes)
The previous project had a lot more structure but also worked well.
Common themes:
- Communication is 2 way
- Both teams are friendly and competent
- Customer care about results and leave programming to us
- Clear communication about what they hope, but without stress. Especially the first project were the stakes were serious: if we manage to hit the deadline we knew we would save the organization millions, but if not, nobody was in trouble. It was an actual challenge, not a scary thing.
Have I seen dysfunctional Scrum and Agile as well? Yes!
Some examples:
- endless estimation meetings which not only eats programmer hours but also mean that everyone feel they have to match the estimates
- one way communication (in a loop from customer - ux - programmer - tester - customer). Doesn't help if there are 14 days sprints when every sprint is a mini waterfall
- taking time of the project to do agile workshop after agile workshop while continuing to be absolutely rigid
- "release" after "release" but no actual customer
- "finish one thing" taken to mean that styling has to be perfect even on placeholder pages
I wonder how much of this is company- or team-dependent. I don't mind sprints, but I also choose what to put on them myself with very rare (like twice-yearly, tops) requests from my manager.
This, it is costly, certain personal walks away. Proper roles seem to emerge anyway.
The cost of control and the costly overhead has everything to do with atomizing the work load. Just like in a distributed system. The synchronization mechanisms get more complicate.
The more layers, the less trustable the organization, since every manager under their manger, is a potentially corruptable.
This makes the communication lines unclear and makes people over promise. Also distributing the workload has diminishing returns.
The language is improving and improving. Some years ago it was way too difficult for the speed gain to be worth it.
But it's become more and more easy to write. Many of the safety arguments Rust has are still technically true, but 90% less true than 6 years ago.
The C++ community is also really friendly and open minded.
It's hard to explain but C++ also has this nice relaxing feel when writing it. Like doing a puzzle. Maybe the cognitive load is very evenly spread? Or the header hpp model forces you to think first in data models and interfaces? I have no idea it's mainly a feeling.
Almost as if people who have a better life are happier, if you believe this study it almost seems like feelings are somehow related to the external world
Nice. there's also a project [0] that uses the RP2040 to talk I2C and present it as USB HID, created in service of the beepberry [1]. Now that I think of it tho this could be made into a very miniature home theater PC remote, a la the Logitech DiNovo
It is though when you realise 60% of their revenue is from people clicking on "nike" to go to "nike.com".
You cannot have a website anymore without paying tribute to gatekeeper Google.
Consumers can switch any time they want, but there is no incentive for them, they don't have to pay when googling for "nike".
But even though there are only 3 companies in the world that build what I build, all those three companies pay thousands of euros a month so users can find them, even on the exact search terms.
I think our numerical system is a bit like Fahrenheit.
Fahrenheit draws a line between three points and assumes a "law in nature". This was false (short answer).
The fact that there are things like PI for a circle means our numerical system has been invented before we had any knowledge about quantum quantities for lack of a better word (like autonomous autos X).
Math is wrong. But it's gotten this "holy" status of almost defining intelligence, like Latin is to English.
It's clearly wrong (to me at least) but it's such a big part of "smart" people's egos.
PI appears when you’re trying to mathematically describe an oscillating quantity. So it could be a circle, or it could be a wave. So I don’t think it’s the red flag you think it is.
Now #1 on HN. Destiny.
reply