dwheeler

83 ms·

871.
by dwheeler 7y ago
The Python 2 -> 3 transition has been unusually bad. In most programming languages, a change to the language is practically always backwards-compatible. The update may add new capabilities, but the old capabilities and syntax remain i
872.
by dwheeler 7y ago
It's cool that this exists. I might even try to port a program to it, for my own amusement. But I don't see the big draw. VMS had its pros and cons as compared to Unix/Linux. VMS' help system is really nice. However, V
873.
by dwheeler 7y ago
One of the improvements is is support for 16-bit float, to help some machine learning use cases. Cool.
874.
by dwheeler 7y ago
Solution: Firefox.
875.
by dwheeler 7y ago
From its home page: "ScummVM is a program which allows you to run certain classic graphical point-and-click adventure games, provided you already have their data files. The clever part about this: ScummVM just replaces the executables
876.
by dwheeler 7y ago
This is a really good post. That said, I would have listed caches first. Caches are the real key to excellent performance in Ruby on Rails.
877.
by dwheeler 7y ago
SPARK is a subset/superset of Ada. That is, the code is compiled using usual Ada rules, but to pass the SPARK verification checks requires passing some additional rules. Ada's access types (pointers) are traditional; to use acce
878.
by dwheeler 7y ago
Comparing these different approaches is not trivial, of course. One view is to look at "Formalizing 100 Theorems" by Freek Wiedijk, which lists 100 mathematical theorems and the various systems that have formalized a nontrivial nu
879.
by dwheeler 7y ago
> It is not difficult to keep the notation sane in formalizations based on set theory. Keeping notation sane is, of course, very important. I think keeping notation sane is harder than it might first appear, but it is certainly possible
880.
by dwheeler 7y ago
> It seems like encoding ‘all of math so far’ in some theorem prover seems like an ideal open source project... Is there such an effort now? Yes. Metamath's set.mm is expressly such a project, and I recently created a Gource visual
881.
by dwheeler 7y ago
> What I was trying to ask was if there are formally provable theories that cannot be expressed in a way that would let a computer "compute" the proof. Do you mean "find the proof" or "verify the proof"? Th
882.
by dwheeler 7y ago
I don't know whether or not Lean is the "best" (no doubt that depends on what aspects you think are important). But I completely agree with the speaker that there is a need to formalize mathematics with computer verification.
883.
by dwheeler 7y ago
There you go. In general Bluetooth provides lower-quality audio. In addition, in my experience Bluetooth is pretty much always laggy, which is a problem for conference calls. Finally, I feel far more confident about the security of wired co
884.
by dwheeler 7y ago
That is true, but I find the less than or equal to be logically clearer. YMMV.
885.
by dwheeler 7y ago
It's true that strlcpy is easier to use. One challenge is that strlcpy always reads its inputs to completion, even when it is not needed to make the copy, because it has to compute the total length of the screen ignoring limits. In som
886.
by dwheeler 7y ago
This construct: while(*p++ = *q++); is simple. But I agree with you, its time has long gone, because in many programs, this code is also wrong. This idiom assumes that that the source can never be longer than the destination. Th
887.
by dwheeler 7y ago
I think memccpy is an improvement, so I support it, but it's still complicated to use in practice. The article gives this example for a copy followed by concatenation: char *p = memccpy (d, s1, '\0', dsize); dsize -
888.
by dwheeler 7y ago
But most people want to use standard, portable calls. The C standard tried to add such functions in "Annex K", but unfortunately annex K hasn't received much of a pickup (for various reasons). So in many places the problem co
889.
by dwheeler 7y ago
memccpy definitely has some advantages, so it definitely should be in the ISO standard. But memccpy has its own problems. In particular, when concatenating you have to constantly recalculate the "space remaining"; that is just a
890.
by dwheeler 7y ago
I agree, this is a site that shouldn't exist, it's security disaster. Duo should blacklist all their numbers. One easy way would be to detect which accounts consistently confirm instantly; since humans can't do that, those
891.
by dwheeler 7y ago
I appreciate nofollow and ugc. I run a site with user-generated content, and worry about spammers over writing the site with junk content. Having a way to indicate that the link is from someone else is useful for eliminating the incentives
892.
by dwheeler 7y ago
> Though the new reaction with flourine was better than "any other calcium based electrolyte" how did it compare to lithium based models? I noticed that omission right away. I'm assuming that it's nowhere near as good
893.
by dwheeler 7y ago
It's probably not realistic for the average Rails consumer to constantly run tests against master. Indeed, that's probably true for every framework, not just Rails. But it's definitely excellent for some sites to test agai
894.
by dwheeler 7y ago
Yes, that's what I meant. It's true that many GNU projects were (and are) cathedrals, but their cathedrals were more like a bazaar compared to the BSDs. GNU projects tend to be run mostly-independently, but the BSDs were (and ar
895.
by dwheeler 7y ago
I meant the USL vs. BSDi case. That case was, of course, focused on the BSDs. While GNU and Linux were independently implemented, I recall there being some claims at the time that the USL vs. BSDi case also implicated GNU and Linux. The S
896.
by dwheeler 7y ago
Linux came after the BSDs, so you would think the BSDs would have won. There are many reasons Linux-based systems are generally much more popular than the BSDs in the server and workstation spaces. Here's why I think that happened:
897.
by dwheeler 7y ago
> Being able to confirm that your specific vote was counted should be considered a basic human right. No, not if that means that many people cannot safely vote. At one time it was considered a right in many jurisdictions that everyone c
898.
by dwheeler 7y ago
Reproducible builds are extremely important. Most people use pre-compiled code so it's important to have a way to verify that the pre-compiled code was generated from the expected source code. Lots more info is here: https:/&#x2F
899.
by dwheeler 7y ago
Floating point is usually deterministic, but from a math point of view it is inexact (in most cases it only approximates the exact answer). In many cases that is fine, which is why it is used, but it is important to remember that.
900.
by dwheeler 7y ago
It depends on how you use it. Size matters. On a site I manage, we send about 64KB of JavaScript and we are working to make that smaller... and the main site functionality works when JavaScript is disabled. In contrast, many sites seem to

More ›