akho

32 ms·

481.
by akho 5y ago
The reason is that POSIX shells are ugly and stupid, and fish syntax fixes (some of the) fundamental errors made in Bourne shell. Things like "${args[@]}" and other quoting and array-expansion syntaxes should not exist. Subshells
482.
by akho 5y ago
Screenshots looks nice. Maybe you should focus more on the differentiating aspects? The note-taking application market is extremely crowded. (which is indicative that the issues are human, rather than technical — people feel the need to rec
483.
by akho 5y ago
I’m fine with “application” and “app” having different meanings. E. g.: there was recently a post here promoting some group making “an app ecosystem for your terminal”. That’s very convenient — you immediately know you need to close the tab
484.
by akho 5y ago
Few of us would call Emacs an “app”. I’m pretty sure they are referring to mobile/Web stuff. You do have to be very careful there…
485.
by akho 5y ago
XMPP is decentralized in practice (as opposed to Matrix, which is decentralized in theory, with everyone using the same client on the same server using the sole available server implementation). There are hundreds of public servers and doze
486.
by akho 5y ago
XMPP has bridges.
487.
by akho 5y ago
Is “sounding millennial” meant as a young people thing, or an old people thing?
488.
by akho 5y ago
Does Python give you an actual productivity edge? I do not see how. Python developers are easier to hire (certainly than Nim), have a monolanguage experience (meaning less onboarding), but are not magically more productive. Developer availa
489.
by akho 5y ago
Nah, it’s just cheaper to print (small type), bind (once, and badly), and to manage inventory in some edge cases.
490.
by akho 5y ago
There seems to be a weird reluctance in the American publishing market to split larger books into volumes. "War and Peace" editions I see on Amazon are typically single-volume; Russian editions are either two or four volumes. Unfo
491.
by akho 5y ago
The government used the ‘extremist’ classification, yes. Telegram’s Durov referred to the ‘silent days’ practice as his excuse.
492.
by akho 5y ago
“One per district” is 225, “all popular” in contested districts is, like, maybe 70? Smart voting does not create new popular candidates. It’s a way to concentrate the 5 % they can access (or 2 %, if badly mismanaged by publishing the list t
493.
by akho 5y ago
The candidates are known beforehand. Putin’s men do not need Navalny to tell them whether there is a popular candidate in an electoral district. Consolidation of “protest” votes does not make someone win an election; it gives a small extra
494.
by akho 5y ago
> It is strictly enforced in France It’s not strictly enforced. Exit polls published outside France are collected inside France. Those who collect with clear intent to immediately publish are subject to French law, and may be fined. Whic
495.
by akho 5y ago
Sure. How does this relate to the very late publishing of lists? Or to complying with “silent days”?
496.
by akho 5y ago
You either believe that this regulation makes sense, or you don’t. France, apparently, is in a situation where nobody cares enough to enforce, but taking the law off the books is not politically feasible. I do not know whether that’s a good
497.
by akho 5y ago
Doing that before the election (I assume your plan, if you were Putin, is to have all 225 “recommended” candidates withdraw?) is more difficult than subverting the, very optimistically, 30 that will get elected.
498.
by akho 5y ago
One Communist Party candidate per electoral district, yes.
499.
by akho 5y ago
The government will have plenty of time to do that after the election.
500.
by akho 5y ago
“Silent days” are still a thing. The new bit is that the day before the election is not included, only the election days themselves. This prohibition makes sense to me. My personal opinion is that telegram’s actions are reasonable in this c
501.
by akho 5y ago
I expect to see lots of new “educational” apps with addictive qualities.
502.
by akho 5y ago
Quicksort is a specific algorithm involving two array pointers and swapping values. The haskell implementation in question is not it. It’s not a “claim”, it is what it is.
503.
by akho 5y ago
The C code implement quicksort, the Haskell code does not. It has different (worse) space characteristics. Proper quicksort in Haskell is not a two-liner.
504.
by akho 5y ago
To be fair, most of the examples in the post are about shitty hardware, or shitty mobile experiences. The option of using baseline commodity hardware with no RGB lights is somehow not considered. Free software is indeed a breath of fresh ai
505.
by akho 5y ago
I’m sorry, could you give some context to non-Americans? What does “taking geometry” mean? My school had geometry as a subject in grades 7-11 (with 11 being the final year), were you expected to pack it all in one? How does that work?
506.
by akho 5y ago
Functionality is part of UX. Also, “good” is a very high (and ambiguous) bar. Can you name any commercial apps you believe to be good?
507.
by akho 5y ago
I just dislike the whole “there’s an app for that” mindset. Adding an app is not a linear increase in complexity, it’s quadratic: you now need to somehow make it play well with everything you already do. I don’t want more apps, I want bette
508.
by akho 5y ago
Last music player UI I liked was Muine circa 2007. I don’t think we are improving our solutions.
509.
by akho 5y ago
The “more friendly” claim needs to be backed up with experience. I don’t think any of the similar products (numi, Calca, …) have reached a wider audience than Org. Excel, of course, continues to dominate the space, despite being much more c
510.
by akho 5y ago
Highly relevant username.

More ›