SloopJon

One of the most recent citations (Shull et al. 2002) is freely downloadable:

https://www.computer.org/csdl/proceedings-article/metrics/20...

Its own citations may not be satisfying, but I find it nevertheless interesting. Here's the summary of the eWorkshop discussion of "Effort to find and fix":

> A 100:1 increase in effort from early phases to post-delivery was a usable heuristic for severe defects, but for non-severe defects the effort increase was not nearly as large. However, this heuristic is appropriate only for certain development models with a clearly defined release point; research has not yet targeted new paradigms such as extreme programming (XP), which has no meaningful distinction between "early" and "late" development phases.

I don't recall encountering this specific 100x claim, but I do recall Steve McConnell making a similar case early in Code Complete. In chapter three, he has a chart showing that the average cost of fixing defects post-release is 10-100 more expensive than during requirements, which he supports with the following citations:

"Design and Code Inspections to Reduce Errors in Program Development" (Fagan 1976)

Software Defect Removal (Dunn 1984)

"Software Process Improvement at Hughes Aircraft" (Humphrey, Snyder, and Willis 1991)

"Calculating the Return on Investment from More Effective Requirements Management" (Leffingwell 1997)

"Hughes Aircraft's Widespread Deployment of a Continuously Improving Software Process" (Willis et al. 1998)

"An Economic Release Decision Model: Insights into Software Project Management" (Grady 1999)

"What We Have Learned About Fighting Defects" (Shull et al. 2002)

Balancing Agility and Discipline: A Guide for the Perplexed (Boehm and Turner 2004)

I only have one of these (Dunn), and it's boxed up in the attic, so I can't readily check their sources, but I somehow doubt that all of these simply launder the study under discussion.

I don't want to trivialize some of the good points that Hillel Wayne makes about soft-science research applied to software productivity, but he would have us dismiss all of these citations out of hand, simply because they predate capital-A Agile, which of course changes everything. That doesn't strike me as a particularly compelling approach either.

You're right. I didn't see that text anywhere on the page, but it indeed comes from the title element. It doesn't fit in the tab, but I see it now in the tooltip when I hover over the tab in Firefox.

The current title is "Our lawsuit against ChessBase – Stockfish – open-source Chess Engine", which I found confusing. Is Stockfish being sued? Does it belong to ChessBase? I suggest sticking with the simpler title from the blog, "Our lawsuit against ChessBase".

How does Markdown fill the niche of TeX or Scientific Word? Is there a mathematically oriented fork? You mentioned R Markdown in another comment, but I don't see anything in the gallery that looks germane.

I said that they should. I don't know whether they do.

A March 2021 tweet indicated that beta access will be as simple as pressing a "Download Beta" button in the car. A subsequent tweet said that the program had 2,000 owners, some of whom have had their access revoked for not paying enough attention to the road.

The Twitter thread is lacking in attributions, but I saw most of these some months back after a post about scary FSD behavior. I watched with morbid curiosity, and a rising level of anger.

The guy with the white ball cap repeatedly attempted a left turn across two lanes of fast-moving traffic, with a drone providing overhead views. He seemed smart, aware, and even a bit conservative in intervening. Nonetheless, I couldn't help but thinking that none of the oncoming vehicles consented to the experiments racking up YouTube views. If he doesn't jump on the brakes at the right time, he potentially causes a head-on collision with a good chance of fatalities.

Yes, I do agree that beta drivers should get extra training. However, I'm not sure I agree with the premise of beta testing an aluminum rocket on wheels on public roads in the first place.

I wrote some Unicode generators for RapidCheck that worked perfectly well with GCC and Clang on Unix, but not at all with Visual C++ on Windows. The idea was to generate arbitrary code points within various ranges (ASCII, Latin-1, BMP, etc. for shrinking) as UTF-32 / UCS-4 in a u32string, then convert to the datatype and encoding for the API under test--UTF-8 in a string, UTF-16 in a u16string, UTF-something in a wstring, etc. The problem was, some of the conversion facets just aren't supported in the Visual C++ runtime library. I think I ended up doing the UTF-32 to UTF-16 conversion myself.

The other thing I ran into recently, is that u"" string literals that worked on GCC / Linux, Clang / Mac, and VC++ 2017 / Windows 2012 did not work for a colleague on VC++ 2019 / Windows 10. An emoji, for example, came out as four char16_t code units (one for each byte of the UTF-8 encoding, I think), instead of two. We ended up using Unicode escapes instead, although the source code is less colorful without a pile of poo.

This ztd.text library looks interesting, although it's a little discouraging that the getting started section of the documentation is empty. Is this a header-only library?

I bought some Corel stock back then, thinking that the company would somehow leverage its stewardship of WordPerfect to Linux dominance. This was back in the days when lawyers didn't trust Microsoft Word to count words reliably.

I eventually cut my losses, selling my $500-ish stake for $90 or so. So much for the year of Linux on the desktop. At least they didn't go the route of Caldera.

I'll give my SSN to a healthcare provider, and maybe a bank. Random vendor using Stripe? Probably not.

Edit: to be a little less flippant, what is an example of a Stripe user to whom you would be comfortable giving your SSN?

This is primarily about movies, with music handled by iCloud Music Library. Several years ago, I ripped my CD collection to whole-album FLAC files with CUE sheets using XLD. This works reasonably well in VOX for Mac and foobar2000 for Windows.

When I looked into media server software, I got a bit overwhelmed. I can't necessarily tell that something doesn't support CUE sheets properly until after I've spent an hour or two setting it up. Can anyone recommend something that will let me browse and stream a big pile of album FLACs by track?

Although Richard wrote crunch in 2009, he didn't publish it until 2011. Therefore, it doesn't serve as prior art to invalidate the patent.

Richard didn't apply for a patent himself, so first-to-invent is not applicable either. If he had learned of Doug's patent application, and then made an application of his own, there would have been a legitimate dispute as to priority. Since Richard shopped the library around in 2009, he might have been able to prove that he conceived the invention first. However, Doug might then have argued that Richard had abandoned, suppressed, or concealed it. See this blog post from 2005 discussing a similar hypothetical:

https://www.irmi.com/articles/expert-commentary/the-dangers-...

The technique employed by crunch was effectively a trade secret until its open source release. It's entirely possible that the patent is invalid for other reasons, but my layman's understanding of patent law is that because Richard didn't publish or patent in time, crunch can in fact infringe Doug's patent.

This is from 2018, with an update from 2019. It primarily focuses on the Wooting one, which is "sold out forever."

The key touted advantage of the Wooting one keyboard seems to be optical switches with an adjustable actuation point. It also uses a multiplexer for each row of keys, which is somehow faster than a scan matrix.

I don't see an actual test result for the keyboard that includes all of the latency factors discussed. The closest is a link to a tweet by Sunjun Kim. Unlike Dan Luu's measurements, which actually pressed a key, Sunjun triggered the optical switch using an LED, resulting in latency of 4.2 to 9.6 ms, depending on the firmware.

According to this post, key travel can add up to 10 ms, which the adjustable actuation is supposed to minimize. That would put the total latency at 10 to 15 ms, which is around what Dan Luu measured for Apple's Magic Keyboard.

A fully split Advantage is a really interesting prospect. I have a Freestyle Edge RGB at home that I really like, and an Advantage at work that I haven't used in over a year.

My biggest complaint about the Advantage is the placement of the arrow keys in the bottom row. I never got used to them (and I sometimes use vi with a Dvorak layout). I hit them by accident far more often than on purpose. From the survey, it appears to me that they are still in approximately the same location.

I don't have a sense of how big a deal this is, but I had a totally random thought I after I read this:

> FDA currently recognizes the high danger of this compound and lists it as a "Class 1 solvent" that "should not be employed in the manufacture of drug substances, excipients, and drug products because of their unacceptable toxicity ... However, if their use is unavoidable in order to produce a drug product with a significant therapeutic advance, then their levels should be restricted" and benzene is restricted under such guidance to 2 parts per million ("ppm").

This reminds me of a story I heard about Kosher Coca-Cola. They had designed it such that its impurities were below a certain threshold. (Googling it now, the ingredient in question was glycerin derived from non-Kosher beef tallow.) However, the consulting rabbi explained that the threshold only applied to accidental impurities; you can't put them in on purpose.

iStat Menus:

> You can tell, since the screenshot of the power draw monitoring utility (iStat Menus) is twice as sharp in the second example.

I got a 16" MBP last year in order to WFH. I could not get it to work with an old 1920x1200 Dell monitor using DVI adapters. Months later I bought a 1440 monitor that works over HDMI and DisplayPort, but the display is fuzzy. I found some tweaks to improve it, but it's still noticeably worse than a hackintosh running High Sierra.

I have felt like the performance isn't great, but I chalked it up to Docker and the Linux VM it uses. However, I also noticed that the computer was warm to the touch with fans running even when I hadn't used it for hours. I've taken to manually forcing it to sleep at the end of the day.

I'm excited to try this HiDPI hack, because it may solve the power usage problem and the fuzzy display. It would be a real shame to lose the built-in display if this only works in clamshell mode, but it just might be worth it.

Edit: I just realized that I can't use Touch ID in clamshell mode. That makes this a harder call.

The types of surprises that I've seen have to do with inferences that the compiler draws: this program would be undefined if foo was negative, therefore foo is not negative, therefore I can optimize away this condition.

Probably because "+" is a special character in most shells. You'd have to quote the filename every time you did anything with it.

Edit: morning brain fart. Actually, I guess "+" is not all that special.

I thought for sure that someone would have posted a link to that xkcd comic by now. I only dabble with higher-level APIs, so I can't judge this on the merits. If NVIDIA really continues to back this, and follows through on wrapping other libraries like cuDNN, it could be a whole new level of vendor lock in as people start writing code that targets CUDA Python. I think the real test will be whether one of the big projects like PyTorch or TensorFlow gets on board.

I've always found the idea of master recordings curious. Once a song becomes a hit, the band may perform it almost every day for the rest of their career, but what we listen to through our headphones or speakers is this one particular moment in time. Yes, there are live recordings and bootlegs, but by and large it's this one master recording, which may get dusted off every couple of decades for a remaster.

Some masters are guarded jealously by their owners. For example, I've heard that some of the Stevie Wonder releases aren't even from the masters, because he won't part with them. Some masters may truly capture something irreproducible; say, John Lennon's hoarse voice in "Twist and Shout."

In this particular case, at least, Taylor Swift has put the lie to the uniqueness of the masters. If she has the right to rerecord everything, the old masters aren't so special. But I think the article is also right. After this exercise, she may have made her point, and may force Shamrock Capital to make a deal.

I've never heard of Cherry switches on a laptop. Apparently this is the first model with Cherry MX Ultra Low Profile switches:

https://www.cherrymx.de/en/blog/cherry-mx-ultra-low-profile....

The paper is here (referenced in the Loom docs.):

http://demsky.eecs.uci.edu/publications/c11modelcheck.pdf

This page describes how to get the code (linked from the paper):

http://plrg.ics.uci.edu/software_page/42-2/

I found some data races in an ODBC driver using TSan. Running it through CDSChecker is one of my round tuit projects.

The Github README shows that the public API uses types from Abseil, a library that "promises no ABI stability, even from one day to the next." That seems problematic.

> You know there are issues, so that is a good thing.

That was my thought too. Programming is kind of like American Idol--a lot of programmers don't know how bad they are. The submitter said they failed to accomplish a medium complexity task, but it's not clear what this means: got stuck and gave up; missed a deadline; finished, but with poor quality?

From one of the submitter's thirty previous Ask HN submissions, I see experience working with a variety of tools, including machine learning. With batteries-included languages and powerful libraries, you can get pretty far in Tinkertoy mode, assembling the pieces of a solution. When you're faced with a task that demands more programming skill, it can be an adjustment.

If you have a passion for software development, self awareness, and a desire to improve, I think you'll be okay.

Interesting coincidence that the earlier submission has 340 points. I'm upvoting this submission for the link to AZdecrypt:

https://www.zodiackillersite.com/viewtopic.php?f=81&t=3198

Although it's a Windows executable, it looks like it's possible to build from the source in this repo:

https://github.com/doranchak/azdecrypt

I haven't seen any pricing on the CMP cards yet. Are you sure they're more expensive?

It is a sad reflection of 2020 that the Ars Technica system guide is a review of three mainstream prebuilt systems. I looked into the Lenovo, HP, and Dell systems, each of which have showed up on deal sites recently with an RTX 3070, 3080, or 3090. Here are some things I found:

One review claimed that the HP Omen has some kind of RAM serial number check in the BIOS: even if you buy the exact brand and model of HyperX Fury RAM, it will run slower, unless you buy it from HP. Generally, the HP Omen 25L and 30L were both knocked for loud/inadequate cooling. There are also complaints about difficulty canceling orders after discovering long shipping times that were not adequately disclosed at order time.

I didn't research the Alienware Aurora as much as the other two, but I did find some complaints about loud/inadequate cooling.

The Lenovo Legion Tower 7i looked really good. Decent design with adequate cooling. Shipping times are reasonable, because they only seem to list configurations that they actually have. If there is a complaint, it's that their configurations are fixed. You can't configure the RAM or storage. I didn't find as much detail on the 5i, but it seems reasonable within its limits (e.g., doesn't seem to be available with RTX 3080).

I also looked at prebuilts from CyberPower, iBuyPower, Corsair, Falcon Northwest, Origin, and others. I presume that their build quality is superior and/or less proprietary, which is generally reflected in their price.

I gave up on building my own this time around, and bought an MSI Trident X that happened to show up in stock one evening at Adorama. Its 1TB SSD came partitioned strangely (split into two big partitions with a recovery partition in between them), but MSI has a video on YouTube demonstrating how to fix that using Partition Assistant. I've only had it a few days, but I'm impressed that they squeezed a 10700K and RTX 3080 into a box significantly smaller and quieter than my home-built Ivy Bridge with a GTX 1060. I was able to add a 2TB M.2 with no trouble, and there's even room for 2.5" drive or two.

> We are speaking about languages, not which implementations are available.

When it comes to adoption, an area where Eiffel has clearly lagged, the availability of implementations can't be an afterthought; not when you have good, free implementations for Python, Ruby, Go, Rust, Julia, and many more.

Objective-C and Swift are interesting languages that are held back by the fact that the best environment is Xcode on a Mac. Similarly, Qt has been plagued by license anxiety for years (first the QPL, and now the delayed LGPL releases). Therefore, I would say that one of the "obvious, non-negotiable things any new OO language should have," to quote the article, is a good open source implementation.

Here's how Eiffel Software describes the licensing for EiffelStudio:

> If you wish to earn a commercial benefit from your application and not release its source code, you must purchase the number of licenses you need for your development from Eiffel Software. After you purchase licenses, you are free to use and distribute your application the way you want (see the end user license agreement).

> If you select the Open Source license, you must release your development under an Open Source license for the benefit of the community at large.

So I'm still confused about the licensing. For the open source license, is a GPL runtime embedded in the compiled program or something?