|
Must be Chrome and Firefox users. Edge users are alerted to this problem so they can change their passwords.
|
|
|
|
|
No, Chrome users are too, but it is more subtle... You have to search it out on your account.
|
|
|
|
|
Intel, TSMC, Samsung, ARM, Qualcomm, and more are coming together to define a new chiplet connectivity standard Great - until you step on your CPU in the dark
|
|
|
|
|
The Data-Oriented Language for Sane Software Development. "I have as many names as there are winds, as many titles as there are ways to die. "
And the language is also metal!
"Odin now officially supports both the Metal API and Direct3D 11 & 12 out-of-the box! This makes Odin currently the only language that officially supports all of the major graphics APIs: OpenGL, Vulkan, Direct3D11, Direct3D12, Metal, and WebGL 1 & 2."
|
|
|
|
|
Hopefully to be relegated to mythology as has the god Odin.
|
|
|
|
|
Dartmouth researchers have built an artificial intelligence model for detecting mental disorders using conversations on Reddit, part of an emerging wave of screening tools that use computers to analyze social media posts and gain an insight into people's mental states. No one run these blurbs through that AI
Seeing as how they trained it on Reddit, I think they'll have a good collection of crazies to choose from.
|
|
|
|
|
Facebook would overwhelm the AI.
|
|
|
|
|
And determines that 99.99% of people using social media have mental disorders.
|
|
|
|
|
To be clear; they trained an "AI" without confirming their conclusion with the actual people. It's useless.
|
|
|
|
|
”A class should only have one reason to change” is a mantra that Object-Oriented advocates have chanted for years. "History doesn't repeat itself, but it often rhymes"
|
|
|
|
|
"A class should have only one reason to change, and that's because I say so!"
|
|
|
|
|
Software development managers can support their most valuable asset—people—with these seven tips. DO: coding katas DON'T: sit-ups
|
|
|
|
|
Security analysts warn of a sharp rise in API attacks over the past year, with most companies still following inadequate practices to tackle the problem. Attack Portal (for) Insecurity
I'm assuming they mean "web endpoints" here when they say APIs
|
|
|
|
|
|
What happens if we save to a standardised text representation, and instead format code on demand, in the editor? "I believe it is peace for our time."
|
|
|
|
|
Kent Sharkey wrote: save to a standardised text representation Ooohhhh!!! Another holy war to set the standard!!!
|
|
|
|
|
|
Why not take it a step further, ALGOL 68 style, defined in abstract tokens. Each implementation may define concrete representations of abstract tokens, e.g. German language keywords. (That has been done!)
The storage format could be a parse tree that could be unparsed at display time, so that block delimiters could be displayed as {} or as BEGIN-END, the datatypes as INT or int or integer, as float or REAL or single, statement ends as newline or with a semicolon. Editing operations could operate directly on the parse tree, rather than on a textualized linear representation that must be parsed again.
If the compiler was integrated into the IDE, the parsing would already be done, so it would save a significant amount of time. And it might be more or less language independent.
I never dug into the gcc suite, so I do not know how much of external representation is lost when you get down to the language independent levels. I guess it would be possible to augment the structures with tags making it possible to reconstruct the source code. Maybe so much of this would be rule based that you could (re?)construct it into a different source language than the original one. Maybe you still need tagging for higher level source construct, but the tags may be general, not tied to one specific source language.
An intermediate language editor that can display the code as Pascal source, C#, Java or C++, or even Fortran IV or Cobol ... That would be really fascinating. Maybe some gcc guru could create something based on that suite's intermediate formats? Or maybe one could go into the Roslyn suite and do something, probably based on an earlier stage than MSIL, but still after tokenizing and parsing, to make it somewhat language independent.
I think this would be far better than sticking to a linear text representation of program source code.
|
|
|
|
|
Can we please have editors that prevent us from entering syntactically incorrect programs in the first place? Imagine the amount of time and mental energy freed up by that, now available to reason about the semantic correctness of our digital utterances instead.
And why, in 2022, do we still stick to editors that display 1D text sequences (files) in a fixed-width grid on 2D displays in what amounts to something like 1.2 dimensions (look up "fractal dimensions") defined by the line breaks in the file? (Rhetorical question.)
We probably won't get in our lifetimes a "PLEASE" command that reads our thoughts and writes our programs for us, and of course we don't want to have to program in machine language, so somewhere in between. If we could start all over, which abstraction level should our programming languages live on, and what would be the best representation for them? Meaning: understanding and reasoning about them?
Can we, should we think even bigger? How would a complete programming system look like that takes advantage of the computing power available in modern machines and adapts a lot more to the way we humans think and work, not just to the bare minimum required to transmit code to the compiler?
|
|
|
|
|
You can already automatically reformat. The so called standardised text is called source code.
Bastard Programmer from Hell
"If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
|
|
|
|
|
I do. Whenever I present code to others, I change the VS Editor options to the current coding standard settings, delete and reinsert the last brace in the file. When I pick up some source file that I will be doing significant work on, it goes the other way: I make sure that the Text Editor options are set to my preferences, delete and reinsert the last brace.
I generally prefer to fit an entire method in the window, so that I do not have to scroll over several window fulls. Besides, when you do not water out the indentation structure with lots of blank lines and brace-only lines, rather letting the indented block visually appear as a block, makes it a lot easier to follow the block structure and program flow. But coding standards don't agree with me. High SLOC is far more essential
|
|
|
|
|
trønderen wrote: I do Once upon a time an employer demanded no formatting, we built SQL as a string without any formatting. So, I made a plugin that did it on demand.
The standard the author wines about is already there, called the syntax. The author doesn't write code. Get off my lawn.
Bastard Programmer from Hell
"If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
|
|
|
|
|
The author has a point in that standard code formatting makes things harder for diff/merge tools. Where they go off the rails is in deciding that the solution is to change how everyone's editor of choice works to make things easier for existing line based diff/merge tools; rather than to make a new diff/merge tool that has a greater awareness of the structure of the code it's examining.
Did you ever see history portrayed as an old man with a wise brow and pulseless heart, weighing all things in the balance of reason?
Is not rather the genius of history like an eternal, imploring maiden, full of fire, with a burning heart and flaming soul, humanly warm and humanly beautiful?
--Zachris Topelius
|
|
|
|
|
Shouldn't a diff/merge tool work on the token level? Pass each candidate file through a tokenizer and compare the token streams.
You could optionally include a parser, recognizing e.g. statements, for reporting multiple (/consecutive) token edits/moves by the next higher unit (statement, at the basic level, but if all tokens within a statement are affected, the same logic may be applied at higher levels).
|
|
|
|
|
As a standard text representation, I suggest the universal language of all non-quantum computers, 1's and 0's!
|
|
|
|