sfvisser

23 ms·

211.
by sfvisser 6y ago
I’m learning rust right now and there is a lot to like. Steady updates like this are also very motivating. The ecosystem feels very sane - especially compared to npm. Top notch Wasm support, cross compiling is a breeze. That said, coming fr
212.
by sfvisser 6y ago
Autonomous robots sound cool, but there are so many problems that could benefit from robotic automation that don’t require full autonomy. In controlled environments where we know the entire state of the (local) world including the position
213.
by sfvisser 6y ago
It makes so much sense for the EU to start investing in an open web and open source software. A relatively cheap option to move towards a society with better data protection and online civil liberties. I’ve never took the time to really dig
214.
by sfvisser 6y ago
Live on earth might actually originate from geothermal vents at extreme (for us) conditions. See the excellent book The Vital Question by Nick Lane.
215.
by sfvisser 6y ago
> SQL is messy because describing the underlying data relationships are messy. Not sure that is true. You can imagine having a smaller query language with clean semantics able to capture messy data relationships. Small functional express
216.
by sfvisser 6y ago
If you’re in management at an engineering dept/co and make decisions about what is going to be engineered and how that’s going to be deployed you are in engineering yourself and should obviously take the oath yourself. Not saying I’
217.
by sfvisser 6y ago
Europe should build their own software. Instead of complaining about US monopolies the EU should stimulate innovative startups right here. Create an environment where talent doesn’t flee to the US. And if it’s not about creating multi-billi
218.
by sfvisser 6y ago
I really don't understand how any of this can be considered scientific. > irreversible collapse of our civilization Strong words. We really have no clue what will happen in the future. Blindly extrapolating current trends to predict
219.
by sfvisser 6y ago
Dependent types strictly just means being able to parametrize types with values. You can design many different types of PLs with this feature. You don’t need strict reasoning for it to be useful like any other type level feature.
220.
by sfvisser 6y ago
From my (admittedly very limited) experience implementing CRDTs it became clear that even technically correct is not good enough. Optimistic merge strategies require a very clear understanding of user intent and expectation. Properly consis
221.
by sfvisser 6y ago
These things are unrelated. Idris is strict and performs heap allocations so does require a GC.
222.
by sfvisser 6y ago
> In a standard rational decision-making framework, if: cost of action < risk, then: take action. Yes, but cost of action includes opportunity cost of not doing all the alternative actions for which you now suddenly don’t have the tim
223.
by sfvisser 6y ago
GHC's type errors can be daunting, but the complexity of the errors usually scales quite linearly with the complexity of your code. That's good, it keeps you in line. > The problem with Haskell is it's purity and thirst fo
224.
by sfvisser 6y ago
It’s obviously ok to be pessimistic, but at least explain why. Do you think the feature is not going to be very useful in a practical setting? Do you fear people are going to use it all over the place in libraries and make life harder for t
225.
by sfvisser 6y ago
> * Error handling always ends up being non-existent or of the quality of "begin, for, if, while, repeat, identifier, number, float expected" with no good way to override what happens Probably true in many toy-versions. However
226.
by sfvisser 6y ago
A 'realistic' drop shadow can never be smaller than the occluding object itself.
227.
by sfvisser 6y ago
If you really need to run your computations in a separate context/arrow, making them explicit in the types is nice. Having general syntactic rules to build, compose and run them: great. However, if you can avoid the additional arrow en
228.
by sfvisser 6y ago
Nice. You can also add a `hue-rotate(180deg)` to somewhat preserve colors and add `img` to invert images back to normal again. I have this as a bookmark and work well in many cases.
229.
by sfvisser 6y ago
Maybe some did. Many also moved because finally they could use a decent statically typed language on the client.
230.
by sfvisser 6y ago
Gets me pretty excited though. MS has a track record of productive dev experience and great tooling ecosystems. I’m actually looking forward playing with this!
231.
by sfvisser 6y ago
I just read this as "within the context of MS research projects — maybe even specifically in the category of programming languages — Bosque has a relatively high breakthrough value". But I might be wrong here.
232.
by sfvisser 6y ago
> The main thing being the meaning of having dependencies for useEffect and other hooks. This is 100% a result of how react (ab)uses the host language semantics to implement their DSL to describe components and has nothing to do with the
233.
by sfvisser 6y ago
> So now there is yet another Clash has been around for a while now.
234.
by sfvisser 6y ago
Sure, but talking about the opportunity cost of the thing you did pick feels backwards to me. Also, prioritization is not only minimizing cost, you’re probably also maximizing reward. So the calculation is more complicated anyways.
235.
by sfvisser 6y ago
> The "opportunity cost" _is_ an explanation and your suggestion of shifting the label "opportunity costs" to apply to something else such as "unhappy users" is just adding to the misinterpretation. I’m appl
236.
by sfvisser 6y ago
I don't get the use of the terminology here. Opportunity cost is the cost associated by not reaping the benefits of an alternative not being picked. This is a cost from the perspective of the party making the choice. So in this story,
237.
by sfvisser 6y ago
> Yeah, I also don't think that's correct, in most cases. Actually, stupidcar assessment is completely correct. > React updates the entire tree internally when anything changes, but actually only changes DOM nodes who&#x27
238.
by sfvisser 6y ago
Not a very compelling article. None of the drawbacks stated are really serious issues, or issues at all. Everyone with a bit of experience will be aware of the possible drawbacks and knows how to avoid this. Newcomers probably have other th
239.
by sfvisser 7y ago
Yes, irreplaceable! Still using GitX mostly for creating partial commits by picking the exact lines/blocks that go into a commit.
240.
by sfvisser 7y ago
Has nothing to do with category theory. Not necessarily at least. Also, there is a strong intuition to free theorems that many people without an FP background will easily understand.

More ›