wesnerm2

> I exactly wrote that Microsoft devices were quite bad as they came too soon. And Apple came at the right moment, when low power chips were powerful enough, touchscreens were precise enough without a pen. And device had acceptable consumption, batteries had decent capacity, lifespan.

The argument is heavily reductionist because it treats technical breakthroughs as passive market timing rather than active, high-risk engineering.

Apple did not succeed by waiting for silicon to become fast enough to brute-force a desktop UI into a handheld. Instead, they reimagined the hardware-software stack from the ground up specifically around human fingertips and GPU acceleration.

Apple incorporated a GPU to accelerate the graphics. Apple acquired Fingerworks in 2005 and built custom chips to make multitouch viable. When Blackberry engineers were in disbelief about battery life, they cracked open an iPhone and found the battery occupied nearly the entire case. Apple also invested in aggressive thermal and power co-design such as power-gating states, dynamic clocks, ambient light sensors, etc.

It's because Apple was vertically oriented, they were able to make the necessary hardware advances.

Nakamura was sued for $100 million by Hans Niemann.

https://www.chess.com/news/view/hans-niemann-lawsuit-dismiss...

Directly from the article:

Those states are Arizona, Colorado, Florida, Georgia, Kentucky, Louisiana, Maine, Massachusetts, Nevada, New Mexico, New York, North Carolina, Pennsylvania, Utah, Virginia and West Virginia, along with the District of Columbia.

MathML is machine-friendly not human-friendly.

Upgrading to Windows 11 is possible from an unsupported Windows 10 machine by adjusting a registry setting and downloading an ISO from Microsoft.

Atkinson's HyperCard was released in 1987, before the widespread adoption of the web. HyperCard introduced concepts like interactive stacks of cards, scripting, and linking, which were later adopted and expanded upon in the web. Robert Cailliau, who assisted Tim Berners-Lee in developing the first web browser, was influenced by HyperCard's hyperlink concept.

I think that you are referring to this, which was posted in the past month.

https://news.ycombinator.com/item?id=43739456

https://the-pocket.github.io/PocketFlow-Tutorial-Codebase-Kn...

> There is no performance benefit that immutable structures offer that cannot be had by mutable ones. It is asinine to assert otherwise, especially when literally every single measurement ever done demonstrates that immutable structures perform orders of magnitude slower.

Copying is free. Comparisons and change detection are much faster. Data-sharing, thread-safety, content-addressing, versioning/persistence have faster and more efficient implementations--often for zero cost. Immutable data structures have more guarantees, which lend themselves to more optimizations.

Chart parsing uses immutable data structures and many other DP algorithms rely on immutability to take an algorithm from exponential running time and space to polynomial running time and space. Git uses content-addressing to implement zero-cost branches, which used to be inefficient in traditional version control systems, which were more imperative.

Linked lists were heavily used in application software before the appearance of standard libraries and Java, which is when dynamically sizable array-based lists become common. There also wasn't a gap between the performance of linked lists and arrays before CPU became significantly faster than RAM.

  Location: Philadelphia, PA and Bellevue, WA
  Remote: Yes
  Willing to relocate: NYC, California
  Technologies: AWS, GCP, Azure. My usual stack is C#/.NET Core, but I also professionally developed with C/C++, Javascript, Python, Java, Clojure, and PHP. Also used Ruby, Go, Scala, F#, Objective C. Primarily backend, but developed desktop, mobile, frontend(Blazor, ASP.NET, React, Astro, AngularJS), backend, and mobile applications.
  Resume: http://wesnermoise.com
  Email: wesnerm yahoo.com

Software developer with business degree and expertise in AI, mathematics and algorithms. Former software design engineer in the Microsoft Excel group, where I developed PivotTables. Primary background in .NET (C#) and UNIX (C++) with eclectic interests in frameworks, languages and platforms. Proficient in systems programming, desktop and mobile application development, and web development (UI and API). Independently developed multiple large applications—over 100K lines of code each. Ranked in 99.9th percentile in competitive programming at HackerRank and CodeChef. Filed patents and presented four times at Microsoft technical conferences.

If it is 23M pixels per lens, that is still more resolution than a smartphone's screen. Each lens is smaller than a smartphone's screen and the resolution is per eye. I wouldn't be surprised if this actually exceeds the eye's ability to perceive pixels.

The difference between a monitor and the lens of a headset. If you look at a 4K monitor up closely within a region of the screen of two inches in radius, you are not seeing 4K in that region. 4K of pixel applies to the whole monitor not to the eye's field of view as it does to a headset.

If you were using the headset as a monitor, you could zoom in on text and the text can effectively have infinite resolution as it scales up into view.

This post is not valid because most companies post jobs in Hackernews after several days. In addition, the is-hiring post opened on April 3, 2023--a little over 24 hours ago.

As other people have noticed, April appeared to be the lowest month in new jobs, but the true numbers won't be evident until the month is over. The ratio will likely fall lower with the reasonable assumption that a greater proportion of people seeking work post earlier in the month than of companies seeking employees.

> If you were starting a startup for example, you wouldn’t need to hire any humans - you would just spin up enough AGI instances and they would design your product, write your code, deploy it, handle your financials, respond to any and all customer interactions, proactively navigate regulations and litigation, and everything else that needs to be done in the management of a business. That is the goalpost for AGI. It’s an artificial human - a human replacement.

I disagree with the premise. A single human isn't likely to be able to perform all these functions. Why do you demand GPT-4 encompass all activities? It is already outperforming most humans in standardized tests that rely only on vision and text. A human needs to trained for these tasks.

It's already a human replacement. OpenAI has already said the GPT-4 "with great impact on functions like support, sales, content moderation, and programming."

> > or the “lazy caterer’s sequence” (1, 2, 4, 7, 11, 16, 22, 29,...), the maximum number of pie pieces that can be achieved by n cuts.

> My brain just did not want to accept "7" as the fourth number in the series.

The sequence that you get from the difference of successive numbers in the list is rather predictable: 1,2,3,4,5,6,7, ...

Former Excel 97-XP developer here. Excel 97 had English language formulas. It automatically detected names in tables; however, it was buggy. The feature was removed in a later version (2007?).

C# had value types and pointers from the very beginning. These are not a recent addition. The standard library does know about them. However, not until C# 2.0, which introduced generics, were collections able to avoid boxing value types.

There are some cases where allocations are made when they could have been avoided. Iterating over a dictionary creates a single IEnumerator object. Async methods, tuples, delegates, and lambda expressions also allocate memory as do literal strings. It is possible to have struct-based iterators and disposers. There are some recently added mitigations such as a ValueTask, ValueTuple, function pointers, ref structs, conversions of literals to read-only spans, that eliminate allocations.

DateTime is a value type and doesn't allocate memory. Getting the current time does not allocate memory.

With the recent additions to ref types and Span<>, C# provides a lot of type-safe ways to avoid garbage collections. You can always use pointers if need be.

Git IS heavily used outside of software engineering.

Marketers and documentation writers use markdown and git for building websites and books.

Data scientists use git for their notebooks.

Operations use git for their deployment scripts.

Git isn't the best for office documents. Revision marking (aka, change tracking), document comparisons, and merging are all available in Office suites. Git isn't really used here because there features predate Git, operated on binary files or complex XML documents (neither of which are conducive to diffs).

Git isn't good for databases or data lakes. The files are huge and typically aren't modified (CSV) or are binary/compressed (Parquet).

Former Excel developer here. In Excel 97, the difference between the values needed to be less than 1e-16 times the magnitudes of the operands to be considered equal. It makes sense since such a small proportion is unlikely to occur by chance through arithmetic without the values being "equal" in the original base.

Fact check: The cost of S3 standard storage per TB is approximately $23 not $153. S3 standard storage, the most expensive kind, is $.021-.023 per GB.

S4 costs about the same as S3 Deep Archive storage.

Update: The article appears to be a joke. That aside, there are no cost savings against S3 Deep Archive; additionally S3 Deep Archive supports restores unlike the write-only S4.

Your proposal will reward people who can find a new equivalent job in six months with a temporary 50% raise or fund people for three months to find another job.

The sounds were loud and lasted several seconds. These were removed because most computing devices are now mobile. The sounds are extremely annoying and disruptive to hear from a stranger's laptop every few minutes during a meeting, a classroom lecture, wifi shop, cubicle environment, auditorium, airplane, etc... They could also be embarrassing to anyone who just revealed that they were not paying attention.

You should not make any assumptions about work in the US. There's many American software companies that are able to fulfill your requirements.

In many tech companies, you have "unlimited" vacation. You can probably take more vacation that you already get in Europe. This is true of my company, mParticle.

I can work 9-to-5 as an employee. That is the culture of the company that I work for.

People can choose to work remotely or in an office.

There is also that background or concurrent GC option in .NET, if you have a lot of threads.

You shouldn't read too much from the benchmarks.

The gRPC libraries are primarily developed by Google. Google pours most of its development resources on its core languages: Java, Javascript, Go, C++, and Python. Google then allocates a developer to port the implementation to other languages and platforms. C++, unlike Java, is not typically used for cloud development, so I don't think a lot of resources were allocated for the language.

You can see how the difference in implementation effort can impact performance by comparing dotnet_grpc (Microsoft's fork) against csharp_grpc (Google's original implementation). There's more than a sevenfold improvement in req/s in Microsoft's implementation in the 1 CPU server case (35070 vs 5337), outperforming nearly all the Java benchmarks.

Also, many of those top performing JVM implementations are the same code running under a different garbage collector. .NET has two GCs each with a parallel option, but we only see one benchmark using likely the slower GC (Workstation GC instead of ServerGC).

Longhorn was a failed version of Windows after Windows XP and before Windows Vista.

Longhorn was planned to be a dramatic follow-up version of Windows XP in which the Windows shell was rewritten on top of .NET with three main pillars: communication (WCF), presentation (WPF), and data (WinFS). WinFS was a new filesystem based on a relational database engine (SQL Server).

Longhorn was similar to a previous failed Windows release (Cairo). Longhorn was built on .NET whereas Cairo was built on COM. Longhorn used a new filesystem based on SQL whereas Cairo had a new object-oriented file store. Longhorn use Avalon for the shell with concepts for people/contacts whereas the Cairo used Outlook. Before it became an Office application, Outlook started life as the new Cairo shell in which emails, calendars, task were integrated into explorer. You could do advanced grouping, tagging, and filtering on files. In fact, the first version of Outlook, Outlook 97, as an Office application allowed you to view the file system just like emails.

Longhorn was several years late before the Windows release was canceled with no clear end date in sight.

A reset was directed to produced an incremental new release of Windows called Windows Vista. It would be purely native and ship in a year and a half.

The .NET dependency "Bet on .NET" was jettisoned, and all Window's dependencies needed to be fully released rather than works in progressed. WCF, WPF were released independently. WinFS was killed because it was too slow.

See https://longhorn.ms/the-reset/

I know of two instances:

1) Facebook had a "View as" bug that allowed anyone to view anyone else's private page as that person.

2) Cambridge Analytica in which Facebook shared private information with third-parties under an NDA.

To be fair, we still don't know the loss of votes through mail-in voting. Mail-in voting is more convenient which would likely increase participation, but missing signatures, naked ballots, late or undelivered mail, etc can prevent those ballots from being counted.

When there is a major difference in voting patterns between the two political parties, then polls may not actually be inaccurate. In this case, Democrats chose mail-in voting and Republicans, by and large, chose in-person after being directed by the president.

Also, some percentage of mail-in ballots may not have reached their destination because of possible election interference through reduced post office efficiency--in particular, fewer mail sorting machines, lower hours, deprioritized mails, etc.

In short, the polls don't take into account election interference in the post office or unsuccessful delivery. If most people from both parties were voting in-person, maybe we might have had more parity between the polls and election results.

> Windows has/had a 255-char maximum length of the total path of a file

The actual path of file in Windows can be practically unlimited, but either requires using special network notation that can exceed 256 characters or relative addresses. Recent versions of Windows includes a setting that removes the limitation in the APIs because of development issues like node nested packages.