It’s time to say it out loud: BUILD LESS SOFTWARE
Centering too much on adding new functionality will break your product, your users, and your team. Maybe your business. 1/10
Thread
Conversation
Replying to
Features are not the sole driver of product value. “Quality” almost never comes from new capabilities — usually adding stuff makes things worse for a time. It comes from the quiet work of tending the garden. 2/10
3
74
701
In the rush to find PMF of course you’re going to build like crazy — you’ve got to cram as many ideas into the product as you can to see what plays. If you’re lucky you have the time to evaluate what’s working as you go. 3/10
2
5
216
From PMF onward the drive to add more will never stop. Users have new needs and your product teams have awesome ideas. You fall in love, and you want to give those users everything, you want to support every amazing opportunity your teams can dream up. 4/10
2
4
187
Gravity will always pull you toward building more. And keep building it fast, of course. Don’t worry, we’ll come back and refactor later. You won’t though — unless you treat code quality and design investment as essential to the success of your product. 5/10
2
32
323
It’s fucking hard. No prospect tells you they’d buy your product if only you’d complete that JS to Typescript conversion, or finish your design system. But I’ve worked in an org that was so mired in debt they couldn’t ship. Literally could not. For two years. 6/10
3
32
349
This doesn’t happen with a bang, but with a whimper. Work to the right ignored or delayed in favor of what’s next in the backlog. Code reviews rushed, or guidance ignored. Tests skipped because they’re broken. IA becoming increasingly byzantine. 7/10
1
13
221
Product leaders, y’all need to deeply give a shit about this. And when you get religion and understand how these problems will harm your users, you need to use whatever capital you have to pull back on adding features, as hard as that will be. 8/10
3
20
261
What harms your users will, in the end, harm your business. If you continue to crank out features without regard for the other dimensions of product health, enduring value will elude you. 9/10
3
27
236
This is not a subtweet! More of a reminder to myself. As I said — it’s fucking hard. Keep trying.
BTW if this resonates — I’m hiring for a director of product at Honeycomb. Come talk to us
9
8
243
Replying to
Almost any successful app becomes ever more annoying because more and more features get in the way of doing what the app was originally designed for, until it becomes unbearable and alternatives need to be found until they are killed the same way.
1
1
24
So wouldn't the solution be to BUILD MORE APPS instead of trying to put every single feature into a single one? ;)
1
4
Replying to
It’s easy to build a lot of software but hard as shit to live with it and make it great
2
2
47
The path of improvement does not get folks promoted, most places.
33
Replying to
Breakneck speed of features is a recipe for disaster. It usually implies desperation. Security, skill level, solid engineering all suffer as is noticeable when things break. People immediately notice garbage software and hate on it. Don't be that one.
13
Replying to
Code is a liability. Too many people coding their way out of problems and not thinking. We have a ways to go yet! Super thread.
3
6
Replying to
But without adding any new functionality or feature, the programmers will be out of job. Bugs will eventually be gone if there's no new code added so no need to hire programmers for debugging. The only ones needed would be devops and sysadmins. Now I feel insecure. :(
6
7
Budget a certain percentage of developer hours (10%, 20%, etc) to code quality/streamlining/remediating technical debt. Sales/product people dipping into that walled off time will metaphorically have their heads put on pikes as a warning to their replacement to respect boundaries
2
1
13
Yep! Tech debt budgets are the way to go.
1
4
Quote Tweet
Elon Volo
@igvolow
·
Management objecting to developers responsibly refactoring is sort of like a homeowner telling a roofing company “25 years ago I paid you $10,000 to put shingles on my new house. So why are you asking me to pay you to put shingles on it again?” twitter.com/alexbunardzic/…
1
Replying to
The biggest challenge seems to be if you’re not adding new features, users are going to say why are they continuing to pay subscription fees or for upgrades. Keep improving software without adding a ton of new features is a tough balance.
2
Replying to
The proverb we’ve used these last 40 years is, “no matter how much it costs, it’s always cheaper to buy the app rather than write the app.” (If that app exists). Still 100% true
2
1
Replying to
Most product features updates overlook the learning curve of the users, far and wide.
It should be deeply considered. The time and hassles of more or better features.
Some are a must, but plan should be in place for transfer of knowledge.
1
Replying to
So true. The redesign of already functionning and successful products seem particularly hard too. "Improving" might mean breaking what made the product a success in the first place
1
Replying to
I don’t know - worked at one of the largest FinTech companies on the planet where our devs spun out 100s features/new apps per year. Only a few stuck but they stuck big, which allowed the lesser sticky apps to still have their home with users.
Replying to
Nice thread there should be a certain ratio of technical debt adjusted in every sprint to keep things stable and moving I think.
Engineering teams can manage and prioritise a backlog of their own. I wonder how do you guys manage it at honeycomb
5
Replying to
I just sent note to a team whose product I use daily & was redesigned where the top 4 work flows went from 1-2 clicks ea to 5+/ea w/ some successive clicks on opposite sides of the page & encourages catastrophic errors. Their response: we like the new look.
15
Show more replies
More Tweets
back in 2021 while working on the threads app i once secretly rebuilt our entire ui stack as a joke to look like a certain mid 90s operating system
0:18
105K views
68
345
4,263
Show this thread
afraid to go to sleep because i’ll wake up to a dozen embarrassing bugs, hot takes, misunderstandings caused by our poor explanations, etc. i should really go to sleep though. in some ways we’ve worked on this release for years and in other ways it’s barely an mvp. feels good tho
35
39
1,225
Show this thread
nerds born before 1990 be like “y u no use IRC >:(”
79
119
1,068
"We went remote and increased salaries.
The cost of losing our best engineers is higher than the cost of adapting to market movements."
- Several startup CTOs
6
102
969
Show this thread
putting spaces in folder names should be the 8th deadly sin
53
59
781
Retire early, retire often
15
27
516
At this point, I've touch hundreds of thousands of computers and not one time has Windows built in troubleshooter worked.
39
41
461
My daughter is going to school in a French braid — braided by me. It’s not perfect. But I’m not good at that kind of thing and I’m really proud so clap for me please.
49
1
446
Damn React is 18 and now all the libertarians are no longer interested in it
4
44
356
Show this thread
pro tip:
git rebase -i --exec "cargo fmt" origin/master
this will run 'cargo fmt' on each commit in your branch, stopping if it reformats things so you can take a peek
8
35
333
Not to get all think-piecey on y’all, but I think part of the reason people (like folks in #techtwitter) turn their professions into personality traits is because capitalism forces us to find value in ourselves through what we do and not who we are.
6
78
322
Show this thread
Wanted: an email client that makes it feel socially acceptable to reply to emails as if you're chatting normally, instead of writing a letter.
Best regards,
David K.
23
6
263
feminine urge to turn everything into a compilers problem
5
28
241
Show this thread
If you don’t know something, say so, especially as senior+.
Encouraging this fantasy that you’re supposed to know everything is false and dangerous for both individuals and work culture.
3
39
228
How to build a SaaS startup solo:
1. Try to look for a solution to automate 1 thing you’re annoyed with
2. Be sad that you can’t find any solution
3. Build MVP for yourself
4. Add payment
5. Share it with others & related communities
6. Fix bugs
7. Add dark mode
8. Launch
26
21
212
half my mentions are people saying 5lbs of cheese is nothing, like we get it, you have kids
17
3
206
Trending now
What’s happening
War in Ukraine
LIVE
War in Ukraine continues as Russia says it will cut back military operations near Kyiv
Trending with Russia
#MoonKnight
Series Premiere Now Streaming on Disney+
Promoted by Marvel Studios
Trending in United States
FACE REVEAL
8,854 Tweets
Entertainment
Last night
A viral image of the altercation between Chris Rock and Will Smith has been digitally altered, fact-checkers report
RSVP Magazine
3 hours ago
Jada Pinkett Smith breaks silence after husband Will slapped Chris Rock following alopecia joke