A tweet thread about what the actual job to be done of Saas software is.
.... You're inheriting broken workflows, fragmented data, regulatory overhead, and edge cases no model has ever seen. The clean demo you showed? That was the easy part."
Topic: edge cas
21 chunks · 19 episodes
3.3x burst in 2024 Q2
?
Peak quarter intensity across the topic's active span. Higher values mean attention was concentrated into a shorter stretch rather than spread evenly over time.
Topic summary
?
A short read on the topic's time range, peak episode, and strongest associations. Use it as the quick orientation before drilling into examples.
- edge cas appears in 21 chunks across 19 episodes, from 2024-01-22 to 2026-04-27.
- Its densest episode is Bits and Bobs 5/13/24 (2024-05-13), with 2 observations on this topic.
- Semantically it travels with combinatorial explosion, smuggled infinity, and possibility space, while by chunk count it sits between early stage and mass market; its yearly rank moved from #21 in 2024 to #117 in 2026.
Over time
?
Raw mentions over time. Use this to see absolute attention, not relative rank among all topics.
Observations
?
The primary evidence view for this topic. Sort it chronologically when you want concrete examples behind the larger pattern.
Collaborative text editing with offline sync is hard.
... state of even the best algorithms as a corrupted data state. There are trivial edge case examples that are obviously, straightforwardly, "wrong" from the end user perspective. This is as much a UX problem as an algorithmic one.
Your personal use case is likely an edge case for the aggregator.
Your personal use case is likely an edge case for the aggregator.
One reason computer systems get more complex: handling the exponential blow up of all edge cases for all users.
...ason computer systems get more complex: handling the exponential blow up of all edge cases for all users. But in a world of infinite software, every bit of software could be perfectly tailored to its use.
WYSIWYG systems almost always have some weird edge cases that are hard for users to reason about.
WYSIWYG systems almost always have some weird edge cases that are hard for users to reason about. The fundamental reason is because the view is a reduced-dimension visual projection of the underlying sema...
It is not possible to fully enumerate in human language all of the edge cases of a real world phenomena.
It is not possible to fully enumerate in human language all of the edge cases of a real world phenomena. You get the logarithmic return for exponential cost curve. This curve collapses under its own weight. Each incremental t...
LLMs are unlike normal programs in that they allow open-ended output.
...ch run to be predictable, and as we evolve the software to handle more and more edge cases, to run reliably and increasingly without need for human intervention. Software, when it's finely tuned to its domain, gets increasingly tightly al...
Someone who embraces complexity realizes the tensions are fundamental and cannot be resolved, only dynamically and contextually balanced.
...h a little deeper, work a little harder, be a little smarter, we can banish the edge cases to ever smaller impacts." But complexity is non linear. You don't chase down details and fix the whole. If you tighten here it squishes out there (...
The "no true Scotsman" phenomena is similar to the coastline paradox.
...d a generic example of any given class. Because everything in the details is an edge case. Everything in reality, as you get closer, is fractally complicated. The closer you look, the more variance and wrinkles you find. The average of th...
Most software for specific needs is very simple.
...xity of software goes up super linearly as it needs to be generalized. Handling edge cases is the hard part of software and the more you generalize a bit of software the more edge cases that have to be handled. Before we had compilers fro...
One-size-fits-all products get increasingly hard to maintain and extend.
...ly be able to reason about. When adding a feature, the number of conditions and edge cases to reason about explodes. The explosion of complexity makes the product hard to understand for any single use case, and also extremely expensive to...
Apple has made a "sustaining innovation" bet on AI.
...osoft did with browsers back in the day. Apple has done the "simply cut out the edge cases" opinionated cut, which creates a radically simpler thing than alternatives that still delivers maybe 80% of the value. But if AI has the potential...
If you handle the extremes the middle takes care of itself.
...dle the extremes the middle takes care of itself. Think of extreme cases not as edge cases, but stress cases.
At the beginning, do the simplest, most constrained thing that might work.
...y core, which means you don't have the ballooning complexity from accommodating edge cases, you simply cut them out of the possibility space. You get less possibility space, but in a trade for them being significantly more likely to be vi...
Solving all of the edge cases causes exponential blowup of scope.
Solving all of the edge cases causes exponential blowup of scope. 20% of the cases cause 80% of the work. Subtraction cannot be done by consensus. When people on a team hear an ...
When designing a system and coming across an edge case, what do you do?
When designing a system and coming across an edge case, what do you do? One option is to add extra complexity to the solution. To absorb the edge case and make it fit. This is an approach that engineerin...
Most syncing contexts have to assume that one peer might be gone for an arbitrarily long time and need to re-merge later.
...y merge peers cleanly in the future If you can simplify that, you cut out a 20% edge case that causes 80% of the work.
The metacrap fallacy is "If everyone would simply put things into the right bucket with the right metadata this would work!"
...ost. Every new rule in the model is new surface area. As you go deeper into the edge cases and fractal wrinkles, they get more and more expensive to formally capture and model, while the net value captured goes up much more slowly. You en...
We often need to capture aspects of the real world in formal, black and white models.
...s for 5 tablespoons of tabasco sauce. Is that reasonable?" would be the kind of edge case that is very expensive to exhaustively capture in a formal rule system but is easy if you can just ask an LLM.
It's well established that humans have a strong anchoring bias.
... it's pretty good in most circumstances, actually! We just tend to focus on the edge cases where it doesn't work very well, like adversarial negotiations.