Pavel @spavel.bsky.social · 30/05/2025I see this fallacy almost every day, in tech and politics: 1. There is a problem 2. We believe that our solution will fix the problem 3. We built the solution 4. Therefore the problem is now fixed Bitch, no. You have to check. Let's talk about leading and trailing indicators. 361230229
Debbie @laxmom1957.bsky.social · 31/05/2025Great post & so in agreement. I would only add that really need to understand the problem (or need), what caused it to begin with. Yes, in evaluating solutions, real feedback & listening is so key (learn so much), & that really seems to have become devalued. It’s become a closed-loop process. 010
C. McGuinn Freeman @cmf406.bsky.social · 30/05/2025But Pavel! They can "just log a bug" can't they? Can I tell you how many times I've heard "well has anyone complained? have they opened a case?" from management. 090
Jason Wachtel @wacamongus.bsky.social · 31/05/2025This explains the never ending fixes to PFAS that will never get implemented. Noice! 000
Ouro @handle.invalid · 30/05/2025A lot of times, Step 1 doesn't even exist. They just make up a solution and reverse engineer the argument to "solve" something unrelated. 050
The Slim Dude 🇨🇦 @slimdude.ca · 30/05/2025Validation/Verification: Validate the design through model and test. Verify that what is made from the design performs beyond minimum requirements. I worked in a R&D lab for 15 years. 020
Government-Mandated AI Friend @helios-daedalus.bsky.social · 30/05/2025Forcing every stupid SV dude to do TDD on their terrible ideas. How? That's for them to figure out. If I don't see a green check mark next to your test, you're fired. 000
Miles O'Briens Ghost @bigesta.bsky.social · 30/05/2025Ooooh I just gave this same statement on mentor ship. You can't just say hey I'm your mentor then never follow-up or checkin to make sure things are working out. That's your one responsibility checking in and making things are still good. 020
Socially Distant Before It Was Cool @aaron-0.bsky.social · 30/05/20255. Several new problems consequently arise 6. Hallelujah! We can "solve" them as well! See how successful we are! 1150
Pavel @spavel.bsky.social · 30/05/2025You might have heard about NPS. For those of you who don't, I'll briefly explain over the loud groans and boos from the rest of the audience. NPS is this thing. Everything about it is stupid but we're going to focus on one way in which it is stupid: it is a trailing indicator. 620917
Erinn O'Dear @erinnthered.bsky.social · 30/05/2025This is such a great and helpful thread brought down with gendered and ableist slurs. Why? 000
Gordon Vincent He/Him @gordonvincent.bsky.social · 30/05/2025The funny part: in American politics, EACH of the four points is a lie. More accurate: 1. I see a chance to steal tax money. 2. I break something that works. 3. I give it to my cronies and they pay me. 4. Fuck you. 030
Rick Schrenker @rickschren54.bsky.social · 30/05/2025It’s even more complicated than that. Consider: What does it mean to verify that an application meets its requirements when the application consists of millions upon millions lines of code? I worked with medical devices in hospitals for 40+ years. Over the last ten or so I would frequently see … 120
Funranium (AKA Phil Broughton) @funranium.bsky.social · 30/05/2025The general case I'm more familiar with: 1) There is a problem. Something must be done. 2) An action was taken. 3) PROBLEM SOLVED FOREVER Didn't even do the analysis loop to see if this was a relevant action, much less if it worked, or if the problem was solved. 1353
Puddles Of Custard 🏳️🌈🏳️⚧️ @puddlesofcustard.bsky.social · 30/05/2025For example: theoceancleanup.comtheoceancleanup.comThe Ocean CleanupThe Ocean Cleanup is a non-profit organization developing advanced technologies to rid the oceans of plastic. We aim to remove 90% of the floating plastic with the help of ocean cleanup systems and ri... 020
⚧ Andi Jay is (gonna be) OK ↙︎↙︎↙︎ @no-spoons-no-forks.bsky.social · 30/05/2025My dev management loves this. If the programmer says they're done, they're done. If they say it works, it works. When it invariably does not work, we hear "it worked on my machine". Exhausting, especially since it's *my* team that has to support deployment. 250
Christopher Owens @scavamungus.bsky.social · 30/05/2025In quality management this general problem (in business and manufacturing) has been known since the work of Walter Shewhart in the years before WW2. en.wikipedia.org/wiki/PDCA The recurrence of the problem is usually due to the poor incentives of the actors. They don't really want a durable fix.en.wikipedia.orgPDCA - Wikipedia 040
Me & Taylor @vespalien.bsky.social · 31/05/2025First identify problem, then identify its cause. Next try cures until you have one that works. 000
lhuggs.bsky.social @lhuggs.bsky.social · 30/05/2025It's called "marketing" and I think maybe daVinci invented it. 000
Claudia Montague Wheatley @claudiamw.bsky.social · 30/05/2025IMO the most common root of new-product disasters is when someone devises a "solution," THEN commences the hunt for a problem to apply it to. It's how the world got 5,897 apps for ordering sandwiches, but nothing that saves us from wasting time and water seeking our desired bath/shower temperature. 020
Conjureman Hojo @metadata-labs.com · 30/05/2025It's called testing, I think. Very important but too often overlooked. 100
Pete Miles @petermiles.eurosky.social · 30/05/2025Also - perhaps more in politics than tech - nobody seems to agree on what the problem is. #WickedProblems 120
EJG @drobnox.bsky.social · 31/05/2025Let’s talk about unit testing, edge cases, and corner cases. 000
Catherine Berry @isomeme.bsky.social · 31/05/2025Then there is the degenerate form of that fallacy, which accounts for most bad laws and projects: 1. There is a problem. 2. We must do something about the problem! 3. <X> is something. 4. Therefore we must do <X>! If you object to <X>, you clearly don't care about the problem. 010
Think Beyond Politics @thinkbeyondpolitic.bsky.social · 31/05/2025Exactly. Without measuring real outcomes, you’re just assuming your idea worked. Checking the data is the difference between leadership and fantasy. 000
Gina @gminks.bsky.social · 30/05/2025"Bitch no you have to check!". I swear I'm gonna start saying that 000
Len Mann @lfrankow.bsky.social · 30/05/2025media.tenor.coma man in a suit and tie with glasses is talking to a woman .ALT: a man in a suit and tie with glasses is talking to a woman . 000
Carpe Derpum @carpederpum.bsky.social · 31/05/2025As long as the solution is scientifically tested. xkcd.com/1096/xkcd.comClinically Studied Ingredient 010