|
While Google Translate is far from perfect, it's still a helpful way to gain information or engage in conversation. How many of the languages were invented by the AI?
I'm shocked they haven't had Cantonese until now.
|
|
|
|
|
|
You'd need 4 shift keys!?
I bet predictive AI would work really well on that. Enough that building it into the keyboard itself would be worthwhile. Kinda like the whole drag-spelling thing for SMS.
|
|
|
|
|
That's kind of how the JIS (Japanese) input works - at least from what I could tell watching a friend typing with it. It guesses at the Kanji variants as you're typing.
TTFN - Kent
|
|
|
|
|
In the mid 1980s (before the proliferation of PCs) I was working in Asia.
I was shown a "long-form" Chinese keyboard for a CRT terminal.
It had a matrix of about 30 x 20 keys, each with (if I recall correctly) up to 16 characters, and a 4x4 array of shift keys.
Expert users were capable of more than single figures of characters-per-MINUTE.
(I believe its heritage was typesetting machines.)
They were just starting to develop constructive entry methods, where the character could be "assembled" from multiple keystrokes on a regular sized keyboard, and they had some of what we would now call predictive text to speed things up.
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
A programming language akin to BASIC (in being more like 'natural' language) might be really really strong for languages where the characters have that semantic difference based on context.
I say that not really knowing much about them.
For the insanity that was APL, it probably did have its strengths though and I can see where that contextual difference is basically a condensed form of syntax for things we do in code. Stuff like scoping and "var" and other things where in 'normal' languages the context is king and just maybe requires different syntaxes to get what you want.
Does anyone know if there are any programming languages which are explicitly non-English-language based?
(I asked but then googled it because I was that curious)
I found that amongst their other transgressions, AAPL aided the Chinese with their own BASIC. (1/2 joke)
调试输出("Hello, world!")
That last symbol means "raise the roof" and I won't hear otherwise. lol.
Wenyan sounds pretty interesting.
Non-English-based programming languages - Wikipedia[^]
|
|
|
|
|
jochance wrote: For the insanity that was APL,
Actually, for many years after its introduction, APL wasn't a programming language at all. It was a matrix notation that math professor Kenneth Iverson used for the blackboard when lecturing matrix math at Harvard. Some of his computer friends suggested that maybe you could have a computer actually interpreting this notation.
For math operations, having a three-character symbol to denote an arbitrary inner product is extremely convenient - far more than a doubly nested loop with braces and loop control variables and termination conditions.
The 'insanity' lies in trying to use a hammer as a screwdriver. APL was never meant for developing text processing systems or virus scanners. (Yet you might be surprised by the variety of applications where APL actually is a suitable tool!)
jochance wrote: Does anyone know if there are any programming languages which are explicitly non-English-language based?
There probably are a lot, but English-speaking programmers won't have heard of them. NIH  [Sorry, David O'Neill...]
The MS Excel macro language was initially non-English based; it used function names defined by the language version of the Excel application creating it. So if you received a spreadsheet with macros from a source in another language speaking community, the spreadsheet was useless. Quite early (I guess it was in the jump from version 2 to 6, but correct me if I am wrong) the symbolic identifiers were internally replaced with binary IDs, to be mapped to the language of the actual user in the UI. (Hear! Hear! ... How many of you can, honest to some deity, say that you never the last ten years wrote a UI related text string into your source code? )
Some languages are explicitly defined by 'abstract' tokens. Mapping these onto concrete tokens is left to separate documents, quite similar to some libraries with 'bindings' to several programming languages. Algol68 was one of those. I have seen Algol68 source code in German. I know enough German to recognize all the reserved words (but not all of the variable names and the comments ).
In my student days, we made a simple Pascal preprocessor, allowing us to program in Norwegian (we even chose the 'New Norwegian' variant, based on West Norway dialects): All the preprocessor did was replacing any occurrence of MEDAN...GJER with WHILE...DO, BOLK with PROCEDURE, OM...SÅ with IF...THEN, BYRJ...STOGG with BEGIN...END and so on. This obviously introduced a whole new set of reserved words, but we required them to be in upper case. Pascal casing for all user defined symbols was as established coding practice, so it caused no real problems.
Symbol recognition of keywords happens in the very, very first part of the compilation, in the scanner / tokenizer. Making a different source code language hardly affects the compilator at all. You may not be able to adapt to an arbitrary language, at least not in a 'natural' way, if the structure of the abstract language has a structure different from the desired source language. E.g. Pascal has a distinct token (THEN) separating a condition and an action; C does not. If the desired language never uses any such separator, then requiring source to have it makes it look artificial. But if the grammar does not specify any separator while the desired language always requires one, a source statement may appear to the reader as incomplete. (Of course you could introduce a keyword that parses to nothing, but I and many others dislike such solutions!)
Religious freedom is the freedom to say that two plus two make five.
|
|
|
|
|
I don't think I recall ever reading that APL origin story, thanks!
|
|
|
|
|
Half of IT professionals believe there are devices connected to their network that they're unaware of, despite nearly 60 percent admitting that insecure devices pose a 'very high' or 'high' risk to their organization. And the other half know there are devices no one thinks about
|
|
|
|
|
Kent Sharkey wrote: And the other half know there are devices no one thinks about
30%...
The other 20% has no ![Elephant | [mastadon]](https://codeproject.global.ssl.fastly.net/script/Forums/Images/elephant.gif) ing clue about what there is in their network
M.D.V. 
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Some seven years in the making, the Eclipse Foundation's Theia IDE project is now generally available, emerging from beta to challenge Microsoft's similar Visual Studio Code editor, with which it shares much tech. Visual Studio Code with the Microsoft filed off
"Theia is built on the same Monaco editor that powers VS Code, and it supports the same Language Server Protocol (LSP) and Debug Adapter Protocol (DAP) that provide IntelliSense code completions, error checking and other features. ... Eclipse Theia IDE also supports the same extensions as VS Code."
|
|
|
|
|
0patch, a service that provides micro security patches without disruptions, announced today that it plans to offer security patches for Windows 10 for at least five years after its official end of life, giving customers a chance to stick to their current devices without significant security compromises. Assuming you trust a non-Microsoft company to patch your Windows
Insert predictable, "could they be worse than Microsoft?" here
|
|
|
|
|
Could they be worse than Microsoft? 
|
|
|
|
|
Since we first introduced Fluid Framework, it has been used extensively both within Microsoft and by external customers to build real-time collaborative experiences. With FF 2, we’re making it even easier, more flexible and intuitive to build real-time collaborative applications. Most frameworks make me seek out fluids
|
|
|
|
|
Mandatory: Unintentional framework | CommitStrip[^]
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
This post was originally triggered – and I choose that word carefully – by a recent experience on a cloud cost-optimisation project. These experiences prompted me to consider how things had changed since I started working in software. Coming up soon: "Is asking silly questions a niche skill?"
I mean, I see the point he's trying to make, but Betteridge is certainly more correct here.
|
|
|
|
|
It's definitely accurate that we've gravitated towards specialization. I know because I've resisted that pretty hard.
The perspective that SQL is niche might just involve how specialized vs generalized things are for the one doing the perceiving.
Somebody has to do it though, and at least for some scenarios it's almost definitely going to be someone who is more developer than DBA. Now it well may be a DBA (in title), the point is that some things require such an intimate knowledge of not only the data/table structure, but also how the clients interact with it.
|
|
|
|
|
Kent Sharkey wrote: "Is asking silly questions a niche skill?"
No way... Have you seen the Q&A? That's not niche, that's mainstream  
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
The large language models that power today’s chatbots like ChatGPT, Gemini, and Claude are immensely powerful generative AI systems, and immensely power-hungry ones to boot. {Insert photo of person with light bulb lighting up above their head{
|
|
|
|
|
Quote: research out of University of California, Santa Cruz has shown that modern LLMs running billions of parameters can operate on just 13 watts of power without a loss in performance. That’s roughly the draw of a 100W light bulb,
Huh? 
"the debugger doesn't tell me anything because this code compiles just fine" - random QA comment
"Facebook is where you tell lies to your friends. Twitter is where you tell the truth to strangers." - chriselst
"I don't drink any more... then again, I don't drink any less." - Mike Mullikins uncle
|
|
|
|
|
I think the author is referring to a 100 watt equivalent LED bulb.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Richard Andrew x64 wrote: equivalent LED
The devil's in the details.
"the debugger doesn't tell me anything because this code compiles just fine" - random QA comment
"Facebook is where you tell lies to your friends. Twitter is where you tell the truth to strangers." - chriselst
"I don't drink any more... then again, I don't drink any less." - Mike Mullikins uncle
|
|
|
|
|
But that's not what the author said. Simple mistakes like this legitimately make you question the entire article.
|
|
|
|
|
Counterpoint... if I go to amazon and type "100W bulb" everything I see is LED equivalents.
|
|
|
|
|
"Our AI is the dimmest bulb in the rack!"
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
