730 karma · joined March 19, 2016
* it breaks native keyboard shortcuts. After disabling the shortcut overrides in settings, "/" is a NOP (which is weird, since disabling the overrides worked in Confluence)
* the markup is non-standard (but I can live with it)
* sometimes it will log me out when I want to post a comment and all of what I wrote in the comment box gets lost
As for performance, it would be cool if every single thing wasn't behind an atomic reference counter, making it slower than even garbage-collected Go: https://media.ccc.de/v/35c3-9670-safe_and_secure_drivers_in_... (the relevant part starts at 33:06).
Those, who do want to reproduce, will also be replaced by the next generations. Natural selection doesn't select individuals, it selects genes. And your genes are well-represented in the rest of the population.
When phones with touch screens entered the market, we'd often put up with the latency of touch interaction, but these were irritating nevertheless. Then the early iPhones showed how low the latency could be and how much more pleasant using it is. iPhones degraded in this regard since then and Android phones didn't catch up even to the current iPhones. And I'll never use an Android, one of the main reasons being exactly this: latency.
This right here made my day. No further comment on the state of technology today needed.
However, ignoring the aspect of software distribution, wouldn't you agree that the approach taken by the Linux desktop today is deficient security-wise? For example, I would like to be able to give mbsync (or Thunderbird or whatever) my IMAP password without giving it to any other program. So I don't want to store it in mbsync's config file in plain text. Neither will I use gnome-keyring (or any other keyring) because it doesn't have any kind of "program authorisation". Any program can just spawn a new "secret-tool" process and get my credentials from gnome-keyring.
I've been thinking for a while about implementing a keyring which runs as a daemon with SUID of a dedicated user and checks which program sends requests to it, using /proc/pid/exe, but I'm not sure if it's a secure source of truth: how e.g. namespaces affect what's visible in /proc/pid/exe. I know you've been developing himitsu[1]. Have you thought about this problem in that context?
Arch Linux is not Gentoo. And AUR is only a secondary method of installing software. So I'm not sure what you mean.
Science advances one funeral at a time. ~ Max Planck
I'd add to that that it's not only science that advances in this way.
Yours truly,
C++ dev at work
PS. HN formatting is a devil.
Metal drivers are buggy as well and it's sometimes even better to use OpenGL:
https://twitter.com/pcwalton/status/1255250372215611397
https://twitter.com/pcwalton/status/1255571085304541184
It is an added value for some, and it is a negative value for others.
Every minute that you're feeling great because of being "social" at work is a minute that someone else dreads, because they're feeling as if they were in a circus and would prefer to reserve being social for people they like spending time with, not the people they are forced to spend their time with.
I recommend the book "Altered Traits". It looks at some benefits of meditation documented using scientific methods. But it also says that it may not be good for everyone. Specifically, it may be dangerous for people struggling with depression.
That's an alternative way of using it. And half the things you have in GUI mode stop working.
> Vim is a GUI too
Are you speaking about GVim? That's still really a terminal with Vim launched. You still can't have e.g. headings in documents rendered at a different font size than the rest of the content. You still have to deal with the nonsense of patched fonts to get nice-looking arrows with powerline. Sounds like TUI to me.
That's just sad.
As to CLI editors... I use cat all the time to make notes. Because it's straightforward. Firing up vim or emacs would trigger a mental context-switch (caused by screen redraw), while cat is a good CLI citizen --- it's the best thing to make quick notes with 0 distractions.
My understanding of the subject is based on a course in mathematical logic which I took at a university. According to the lecturer, there are no good books on the subject. There was one book they referenced, but it was fat and unapproachable. So, unfortunately, I can't really give any recommendations.
Most code doesn't check whether incrementing a variable causes an overflow, so in practice the test you're referring to is still vulnerable.
> Assuming The Natural Numbers are consistent then all true statements are provable in some axiom system. Just take the collection of all true statements as the axiom system. Now every true statement is trivially provable.
That axiom system wouldn't be particularly useful to humans though. When we talk about sets of axioms, we almost always talk about finite sets of axioms. This is what makes them useful to us, allows us to use them for describing things.
But you do have the right intuition here. The next step is using the compactness theorem[1].
[1]: On a second thought, let me rephrase that: we have statements about structures partially described by linear algebra which (in ZFC) we can neither prove nor disprove for all of them at the same time.
The whole deal with undecidable statements in mathematics is that in our language we make the illusion that there is only one structure deserving of the name "natural numbers", but "natural numbers" are defined by a set of axioms. What Gödel proved is that, when a set of axioms (and the language that is used) is powerful enough, then this set of axioms is either inconsistent (i.e. it has no model), or there are multiple models and there exist statements in the language which are true in some models, but not in others; so you could say that the "truth status" of these statements isn't decided by the set of axioms.
EDIT: Related issue: given a class of structures, in general it may not be possible to write down a set of axioms for which this class of structures will be the class of all models of this set of axioms. An example of that in first-order logic is well-ordered sets. You need second-order logic for that (i.e. you need to be able to quantify over subsets, instead of just elements of universum).
So the way I think about all that is that sets of axioms are inherently imprecise. When you add another axiom, in order to restrict yourself to a smaller number of structures, you always jump over several of them. You're never able to throw out just one.
In fact, we do know that any statement in linear algebra that is true[1], is provable. That's because vector spaces are first-order structures, so that's covered by Godel's completeness theorem[2].
[1]: I.e. it's true for all models of linear algebra.
[2]: https://en.wikipedia.org/wiki/G%C3%B6del%27s_completeness_th...