norswap

38 ms·

391.
Book Review: The Secret of Our Success(slatestarcodex.com)
2 points by norswap 7y ago|0 comments
392.
by norswap 7y ago
I don't like these posts (for any language, not only Haskell). Invariably they will list many points, most of which are subjective or tangential. There are one or two core points (here I guess it is "Control Flow") that would
393.
by norswap 7y ago
This is trivial to do in most languages (including Java and Go) if you just stop to think about it for a couple of minutes. Just pass some kind of context object down the call stack (or make it available as a global — make it thread-local i
394.
by norswap 7y ago
I'm genuinely curious: do you believe this post in particular to be controversial? I can't picture many people (I can picture some) who would think be in favor of what is described here.
395.
by norswap 7y ago
It's generally better designed, but from own experience, harder to actually get something done with unless you are going to learn it from first principles — which hardly anyone does because while simpler than LaTeX it's still has
396.
by norswap 7y ago
I hope it won't have two different and overlaping control panels this time.
397.
by norswap 7y ago
I think the point is precisely that this is much more declarative than something like Tikz.
398.
by norswap 7y ago
Isn't Usenet even more fragile in terms of centralization and legality? I actually wonder why it hasn't been targeted more? Is there any reason? Or is it just that it is relatively obscure compared to BitTorrent?
399.
by norswap 7y ago
I use torrents a lot, but it's not a perfect solution. For one, P2P is mostly a myth — most people (myself included, I will admit) are pure leechers that only seed at very low bandwidth while downloading. Seeders are either altruistic
400.
by norswap 7y ago
People say "he's from the street" but damn, he's also a celebrity — that's bound to attract a slew of weird and ill-intentioned people. That being said, I doubt he would have shot someone unprovoked. That part is fo
401.
by norswap 7y ago
Well, one of my point is precisely that involving programmers in the documentation process (maybe not force them to write the documentation — but review it and run a feedback loop) is crucial in having good documentation, and actually bet
402.
by norswap 7y ago
Writing good documentation is super hard. I try to do it for my parser ( https://github.com/norswap/autumn4/ ) which is < 10k lines of code — both complete and useful Javadoc and a user guide. It's a lot of
403.
by norswap 7y ago
The conclusion is where the gold is at: > One of the most useful things to come out of mystudy is a collection of the rules of thumb my friends use in their decision making. For example, one of my Ph.D. advisers, Fred Mosteller, told me,
404.
by norswap 7y ago
Wrong, I think white-hat SEO is in making sites that do rank well in search-engines because it fits the criteria the search engines have outlined for sites that are user-friendly: clear outlines, no usability-killing scripts, friendly urls,
405.
by norswap 7y ago
I agree with the author the decline of the humanities is about a narrative winning out on the other. But what about what is actually desirable? Is going to the university to learn something that is only indirectly useful as a job skill some
406.
by norswap 7y ago
There is nothing that wrong with Java 8, honestly. Have we done better since? Debatable, but maybe. But there certainly hasn't been a quantum leap. Two step forwards and one step back, more like (and sometimes the reverse). Just the ti
407.
by norswap 7y ago
Any time the component needs to call a method or access data from the aggregate. The kind of thing that happens all the time but the pattern is abstract so there's no killer example. e.g. the component manages some kind of list, and th
408.
by norswap 7y ago
Haha, I hesitated to add this precision, I just knew someone was going to make this point. Yep, and sometimes it's worth it to do it like that if only to cleanly separate concerns.
409.
by norswap 7y ago
I mostly agree with your point on RAII. But I don't think OP was advocating RAII. The typical argument for GC is that it is less error-prone than even RAII. And I agree with that point. It's a trade-off on which you have to decide
410.
by norswap 7y ago
I've always seen automatic reference counting classified as GC. e.g. https://users.cecs.anu.edu.au/~steveb/pubs/papers/rc-ismm-20... > Reference counting and tracing are the two fundamental approaches
411.
by norswap 7y ago
As a counterpoint, if you give users control, you don't have to figure exactly what they want. They can also get creative. And in some cases that is exactly the right approach.
412.
by norswap 7y ago
This is missing the biggest difference between the two, at least according to me. In composition (and its on-steroid cousin, delegation), the component/delegate can't "call back" into the aggregating object. Of course, t
413.
by norswap 7y ago
Ridiculous. The problem is not that you don't know when/where the lifetime will end — that can usually be characterized by a terse "English" description. The problem is that this lifetime is dynamic in nature. The end
414.
by norswap 7y ago
The world needed a new Duke Nukem Forever, and it got one. The cycle must continue.
415.
by norswap 7y ago
I'm part of those who don't really value "fighting for freedom via using IRC" so I use Slack & Discord. I have no stakes in saving IRC. Also, building and maintaining a free IRCCloud clone is worth much more of 5$&#x
416.
by norswap 7y ago
I never see addressed the biggest reason (according to me) why IRC loses, so let's spell it out. IRC isn't persistent (i.e. you can't read messages delivered while you were offline) unless you setup an IRC shell. This is tech
417.
by norswap 7y ago
For those that are interested in going further, I've written on the biggest teaching/insights of Charlie Munger here: https://norswap.com/munger/ In particular, here is a review/summary of his full list
418.
by norswap 8y ago
Really interesting comment! I tried to use JavaFX a while back and was really put off by it (whereas I've had good experiences in Swing). Some issues (it's been some time so memory is a bit hazy): - Lack of good documentation, bot
419.
by norswap 8y ago
Small question: how can they know that it's a mispredicted branch that is being executed (so that they can set the poison appropriately)? If they have this knowledge, couldn't they simply drop the execution since it's a mispr
420.
by norswap 8y ago
How incredibly sad that this is the top comment at the time of writing.

More ›