SoylentNews is people
SoylentNews is powered by your submissions, so send in your scoop. Only 14 submissions in the queue.
SoylentNews
Will we ever have enough programming languages?
| 4% | 5 votes |
| 32% | 35 votes |
| 13% | 15 votes |
| 34% | 37 votes |
| 3% | 4 votes |
| 6% | 7 votes |
| 4% | 5 votes |
[ Voting Booth | Other Polls | Back Home ]
- Don't complain about lack of options. You've got to pick a few when you do multiple choice. Those are the breaks.
- Feel free to suggest poll ideas if you're feeling creative. I'd strongly suggest reading the past polls first.
- This whole thing is wildly inaccurate. Rounding errors, ballot stuffers, dynamic IPs, firewalls. If you're using these numbers to do anything important, you're insane.
This discussion has been archived.
No new comments can be posted.
Will we ever have enough programming languages?
|
Log In/Create an Account
| Top
| 102 comments
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
(1)
Once you've seen one nuclear war, you've seen them all.
(Score: 4, Funny) by Samantha Wright on Monday March 11, @02:25AM (3 children)
From the makers of other great poll topics, like, "Will we ever have enough opinions?" and "What is your favorite security question?"
(Score: 2) by Fluffeh on Thursday March 14, @12:20AM (1 child)
I feel this is rather on topic:
https://xkcd.com/927/ [xkcd.com]
(Score: 4, Insightful) by Freeman on Tuesday March 19, @03:27PM
It's definitely on topic. I'm thinking that computer design and OSes will change in unknowable ways in the future. Therefore, we will never have "enough" programming languages. There will be enough changes in the future that we can't know that Java/C#/C++/Python/Ruby/BASIC/Fortran will be with us for the long haul or that a new programming language won't come to be that is definitively better than the rest.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 0) by Anonymous Coward on Saturday March 23, @05:48PM
Spit or swallow?
Correct answer: gargle.
(Score: 3, Insightful) by Anonymous Coward on Monday March 11, @02:38AM (2 children)
The answer to this will always depend on the context. In some areas, we already have enough programming languages. In some areas, there are too many. In some areas, there will never be enough. In some areas, there will be languages to simplify things. In some areas, there will be languages to make things lower level. Languages are a tool and what tool is the right tool as well as where there are too many or too few often depends on who is using them and on what.
(Score: 4, Interesting) by zocalo on Monday March 11, @11:48AM (1 child)
We're never going to stop producing new programming languages, and there's a genuine need to do so, even if that need is "for educational purposes", so I guess the actual answer is "No", although my gut (and vote) is saying "Enough already!"
UNIX? They're not even circumcised! Savages!
(Score: 2) by JoeMerchant on Monday March 18, @04:16PM
>or develop a fully-functional language optimised to solve a specific niche problem in a given field for which existing options may not be ideal.
Back before Python got big, a small group of enthusiasts were developing Squirrel Script, apparently they still are: http://www.squirrel-lang.org/ [squirrel-lang.org] Squirrel 3.2 stable Released February 10, 2022 - albeit slowly. I encountered the aftermath of one of those developers in 2013. I would characterize it as a classic case of the "sunk cost fallacy" - they had a big pile of Squirrel Script that really should have been translated to something else in 2009 or before, but they just kept with it. Realistically, it was probably a bit of job security as well. The manager who interviewed me had a check-list of "languages you have used" including Squirrel on it, he said in five years of interviewing nobody had ever checked the box.
I would say, at one point Squirrel was solving problems that Python couldn't, but that was a long long time ago. My proposal, that I was hired to do, was to rewrite the Squirrel in Qt/C++. Five months in the CEO had a crisis of confidence after some bad news from sales. Later that week I got a lead on a better offer elsewhere. Six months in I turned in my two weeks notice - best job change I ever made. That Squirrel based company did not survive COVID.
🌻🌻 [google.com]
(Score: 0, Disagree) by Anonymous Coward on Monday March 11, @05:35AM (1 child)
good programming languages reflect the underlying hardware.
the underlying hardware will always evolve.
so there will always be a need to update hardware-specific languages (think CUDA).
if you ask "will humans write in new languages"?
not sure. do we count compiler writers?
obviously higher level stuff like OpenMP and std::parallel will change less and less after some point.
besides, it looks like LLMs do point the way towards a "do what I mean, not what I say" programming model, but it's questionable whether the hardware documentation will be enough for them to learn new hardware (rather than millions of examples written by humans).
(Score: 3, Insightful) by DannyB on Monday March 11, @03:02PM
Hardware design can change, sometimes radically, to fit important problems.
See: The Connection Machine. Or the Cray One. Some of the first massively parallel machines that were programmed differently to get the most out of them. I remember a news article when the National Weather Service got a Cray decades ago and thus began a trend of newer better forecasting models.
GPUs are but one example of hardware changing to fit the problem.
Now with AI being all the rage we have TPUs (Tensor Processing Units) to do "linear algebra" in higher dimensions. (Tensor is a matrix in more than two dimensions.) These have important applications in simulating people's voices or nude likeness. More research will continue in these areas.
For some of these important applications, hardware organization might change radically. And of course, programming these will be a challenge. So as you say, GOOD programming languages will reflect the underlying hardware.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 4, Insightful) by DannyB on Monday March 11, @02:54PM (13 children)
New languages are grate if they improve the state of the art in some important way. Decades ago this was the reason to create new languages. Thus we were blessed with COBOL. So maybe it matters in what way you intend to improve the state of the art. Or, if there can be only one, like PL/1 it needs to be on all systems and ideal for all uses.
Existing languages can sometimes be extended or improved, like C++ and Object Pascal. Who could possibly think that C++ is too complex or is difficult to write a parser for. Of course C++ templates should be Turing complete [wordpress.com] at compile time!
Then there are the people who think that the language which is good for their own part of the programming universe means that their language must be good for everything else and therefore should be the only one. Yes, you can write anything in C. But that also applies for assembly language. And there is Greenspun's Tenth Rule [wikipedia.org]
If you work on something much higher level and abstract, maybe your language should be abstracted far away from the machine hardware. Your language's notion of the maximum size of an integer should not have any relation whatsoever to the underlying word size of the hardware it runs on. Symbolic manipulations and expression rewriting should be as easy as addition. Some languages might be more declarative than imperative. But you wouldn't use one of these languages to write a device driver, boot loader, kernel module, or micro controller firmware.
Another thing I wish more programmers recognized. If some language, whether I happen to like it or not, is popular, then I recognize it is popular for a reason. Clearly some significant community uses and maintains that language. It must be doing something right for someone.
"Oh, but!!! Some languages are less efficient, or garbage collection is bad!" That is an economic consideration not a technical consideration. It may be far more economical to use that language (thus it actually is even more efficient using some other measurement) because it is more profitable. Clue: long ago computers were very expensive and programmers were cheap. Today programmers are very expensive and nobody counts bytes and machine cycles any Moore. Just throw an extra 64 GB of memory into the server and fill a few of the empty sockets on the motherboard with some extra $15,000 cpu chips.
Finally, if there were (or even could be) one perfect language that was ideal for all possible uses, then we would all be using it already. But I think that can never be. The world needs very high level abstract languages, and we need very low level languages where you can see every clock cycle, every pointer dereference right in the code, and without safety features.
That was my inspiration for this pole.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 0) by Anonymous Coward on Tuesday March 12, @09:05AM
Poles and moores are tropes.
(Score: 2) by Rich on Tuesday March 12, @01:31PM (9 children)
"Programming languages are for humans", humans start counting at one, and therefore arrays have to start at one. That's what they (I think it was Larry Tesler) said and implemented MacApp (the original object-oriented framework) that way. A project I maintain still suffers from that. ;) I keep thinking on how to sort that out at some point in the future, some "Option Base" logic will work, like in ancient BASIC dialects, but I'd rather have otherwise low-intrusive compile-time checks.
(Score: 4, Funny) by DannyB on Tuesday March 12, @01:39PM (5 children)
So much strife and bickering over whether arrays start at zero or one. In the spirit of reconciliation, I propose a reasonable, sane and practical compromise.
Henceforth arrays shall start at 0.5.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 3, Interesting) by Rich on Tuesday March 12, @02:53PM (4 children)
Haha! Welcome to the world of raster graphics. :)
(Some graphics libraries define the integer positions ON the pixel, and to draw a non-aliased 1-pixel rectangle, you have to position it at -0.5/+0.5, whereas QuickDraw has the integer positions BETWEEN the pixels, so the rectangle is at 0/1. ...)
(Score: 3, Interesting) by DannyB on Tuesday March 12, @04:31PM (3 children)
I am VERY familiar with QuickDraw. It was genius.
First, the whole idea that the coordinate system is BETWEEN the pixels because the pixels have non-zero size and the coordinate lines are infinitely thin.
And the regions. A fellow programmer and I puzzled over that for a long time. Until we actually went to the trouble to read out all of the integers of a region and gradually recognize how it was organized. Again genius. Any particular type of raster operation that did scanning, could do it in terms of regions. So clipping was natural.
I fondly remember TMON. Once I had that, I seldom used MacsBug.
I don't know if you remember Timbuktu. Myself and one other built that.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 2) by Rich on Tuesday March 12, @08:45PM (2 children)
The original, yes. Everything after, not so much. Palettes with strange index numbers and an off-by-one count, grayishTextOr not recording to PICT, all that breakage here and there...
This is weird that it's only ever been seen there, despite being a general 2D Set data structure. They did have that "inversion point" patent, but there weren't any workarounds (like a 2D-RLE structure with values instead of inversions) and even after it expired, I haven't seen that structure. Only awkward "List of Rectangle" stabs at the task. I wonder if the concept itself is slightly over the head of the majority of software architects everywhere, or what keeps it from being widely available in libraries. I had to implement our own Region logic (32 bit this time, no more 64K "excel row limit") for the MacApp like framework I maintain, I think when we went to 64, and 32-bit Carbon wasn't available anymore. We don't even use it for graphics much (it's still way better than what Cocoa has, btw, and they even broke some of the inferior stuff in Sonoma), and I could have cheated around that - but the grid views need it for selections! I have no clue how anyone could sensibly do a spreadsheet with selectable cells without that Region data type.
Brain Teaser: How are "Grow", "Shrink", or "Frame" done on Regions ;)
By name, not by function. They must have used it a LOT on remote support for all the creative agencies in the 90s, but I didn't have any remote help desk duties. Tapping into QD, and Color QD no less, in a way to intercept what to relay was black magic back then. Radius with their external displays had a monopoly on that for a while, and that was because some original QD developers worked for them.
(Score: 2) by DannyB on Tuesday March 12, @10:05PM (1 child)
I believe, but do not know for certain, that it was done by offsetting the region by 1 or (other number of) pixel in all 4 directions and union-ing those together. Similarly shrinking would offset them inwardly and intersect them together.
It was my first significant venture into using any 68000 assembly, which was a pleasure to use compared to Intel x86.
Three products Timbuktu, Timbuktu/Remote and Screen Recorder had to patch every QuickDraw call (including color QuickDraw in 1987 with Mac II) and be able to relay it over the network, or modem, or to a file (respectively). When not recording or transmitting the screen, those products' patches to QD traps needed to have the lowest possible addition of latency to the normal operation of the system. I managed to get this close to a single JMP instruction per QD call. We patched the address of every QD call in the trap table to point to our routines. If not active the first instruction was a JMP to the original, there was something conditional about it, but memory fades after so many years. Otherwise a tiny bit of "glue" code vectored it to a Pascal function which captured all the arguments, and any other environment information, such as the active GrafPort, and relayed it over the network, modem or into a file recording, for playback at the other end (or later from the file).
So how did we get Mac Paint and HyperCard to work? Those wrote directly to the screen.
When recording we allocated an array of integers, one per row of the display raster. We scanned the display, slowly over a couple seconds, row by row, calculating a simple checksum of each row and comparing it to the integer we stored for that row. If there was a difference, then something had written pixels directly to the screen and this would need to be copied to the remote end. The details are a bit fuzzy. But this did work and took care of apps that cheated by writing to the screen not using QuickDraw.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 2) by Rich on Tuesday March 12, @11:45PM
The overlay won't work for larger offsets (unless done really slow by stepping one-by-one), and the task to offset the horizontal lines is also hard, a naive approach will get O(n^2) or worse, while AFAIK they are in O(n log n) territory. To my knowledge, they use an extremely unintuitive scheme of flipping the X/Y axis somehow and then sorting the inversion points with an ordinary sort algorithm to only do the vertical lines, but the fact that I can't exactly imagine what they might have been doing elevates that to a seriously high tier level.
(Score: 2) by DannyB on Tuesday March 12, @01:47PM
I had just begun to use MacApp by the time I started using MetroWerks CodeWarrior which had C++ and Java. On the C++ side, it had it's own GUI framework called PowerPlant. I really liked that, I think better than MacApp. It was my first taste of Java. But for reasons I ended up having to use Visual FoxPro to build frameworks and parts of applications. A few years later as the Web was clearly going to be the future, I turned my eye back to Java.
With respect to Larry Tesler, I think he should have stuck with arrays starting with zero and ending at size minus one. Humans may count from one, but programmers almost universally count from zero. And for practical reasons.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 1, Informative) by Anonymous Coward on Tuesday March 12, @02:31PM (1 child)
Humans count from one *verbally*, but when I was kindergarten, the number line we used had a 0 at the beginning.
(Score: 2) by Rich on Tuesday March 12, @03:11PM
Apparently, Mr. Tesler's kindergarten's number lines did NOT have a 0 at the beginning, and now I have to muse that a new TDynamicArray might default to whatever MASetDynamicArrayBase() has installed, with legacy code starting at MASetDynamicArrayBase(1), then transitioning to MASetDynamicArrayBase( kDynamicArrayBaseNone ), where runtime checks will shake out all arrays that haven't been hand-treated, and finally switching to MASetDynamicArrayBase(0).
I'd just prefer the runtime checks to happen at compile time, so maybe a few defines to transition "At()" -> "At1()" -> "At0() + At1()" -> "At1()" -> "At()" might be in order. Or call At1 different, like At"ZeroBased"Index() ... whatever..).
But still, the whole show (classic "C with Classes", as they say) is less of a mess than having to deal with "modern C++".
(Score: 2) by turgid on Sunday March 17, @03:03PM (1 child)
Today programmers are very expensive and nobody counts bytes and machine cycles any Moore.
You'd be surprised in the embedded world.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 2) by janrinok on Sunday March 17, @06:06PM
I agree, but you are looking at a rather niche field. Most programmers do not work on embedded systems of the type that perhaps you do.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 0) by Anonymous Coward on Monday March 11, @05:10PM (2 children)
Languages may eventually reach a state of perfection, but compilers will continue to evolve. If the language can represent all possibilities of present and future hardware without making things difficult for the programmer, then only the compiler needs to be advanced for new hardware features and greater performance. It's very hard, but progress is steadily being made on proving correctness being source and output. There will likely never be only one programming language, as different users have different needs and requirements, but we may eventually reach the point of having only one compiler (and consolidation in that regard has been taking place).
(Score: 2) by DannyB on Wednesday March 13, @07:01PM (1 child)
If languages reach a state of perfection, then I predict hardware will begin to evolve to fit that language. Hardware will be specifically designed to execute that one perfect language better and faster.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 1, Interesting) by Anonymous Coward on Thursday March 14, @04:49AM
We already have hardware evolving to fit programming languages. The best example of that now is that most hardware today has first-class support for call stacks despite being register machines. There has been more explicit examples of hardware evolving to fit the language. It was common back in the day because hardware could evolve faster than software. Java processors are common in a number of fields running on FPGAs. GPGPUs have evolved to meet the language needs by adding instructions and flexibility. The real issue with why it has become rarer today is that hardware used to evolve faster than the languages and their implementations. Now it is the other way around as computer science keeps pushing the boundaries of languages and implementations.
(Score: 3, Funny) by Anonymous Coward on Monday March 11, @11:55PM (1 child)
Alan Turing called. He needs his infinite roll of tape back.
(Score: 3, Insightful) by shrewdsheep on Tuesday March 12, @09:32AM
... and let's go to Church [wikipedia.org]
(Score: 4, Insightful) by Anonymous Coward on Tuesday March 12, @06:26AM (12 children)
There can only be one. All languages other than C are an attempt to recreate C... poorly.
Except FORTH. I'll always have a warm place in my heart for FORTH.
(Score: 3, Insightful) by DannyB on Tuesday March 12, @04:33PM (9 children)
Funny, in high school, I heard from a classmate, who heard from his uncle that all other languages were a subset of FORTRAN IV.
At that time I was using BASIC. I didn't yet know how many things I didn't know. I thought I could do anything in BASIC given enough time.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 2) by istartedi on Wednesday March 13, @12:21AM (6 children)
So did I. That's why I learned assembly.
Appended to the end of comments you post. Max: 120 chars.
(Score: 2) by DannyB on Wednesday March 13, @01:46PM (5 children)
As I matured into an adult, I recognized that the "given enough time" was the catch. Time is money.
Computers got faster, programmers got more expensive, completely flipping the economics of computers being vastly expensive and programmers cheap.
Now development costs and maintenance costs are king. And that is worth spending some cpu cycles and bytes to reduce.
I've only ever used three assemblers. (1) an obscure minicomputer in college. I did write what, at the time, was considered a decent sized program. It gathered info from several sources and produced a list of all the interactive terminals, who was logged in, when, and some other details. (2) Intel 8086 on IBM PC because the BIOS was way too slow at writing to the screen. (3) Motorola 68000 on Macintosh to patch all of the QuickDraw traps in order to implement what probably was the first ever GUI screen sharing program.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 2) by istartedi on Wednesday March 13, @05:08PM (1 child)
I was joking about the speed of BASIC vs. assembly back then.
I only learned the 6502. I was exposed to a little bit of x86 in school, and it soon became readily apparent that for most of us it was better to leave it up to the compiler. It's a complex instruction set. Accent on complex.
The "given enough time" quip also extends to waiting around for CPUs to get faster. No doubt CPUs are fast enough to run the shoot 'em up game I was trying to write in BASIC, although that's out of fashion. Such things are commonly written in other high level languages now, which wrap libraries most likely written in C, which wraps the machine code of course. The days when guys like John Abrash counted cycles are *mostly* behind us. I never got that deep in to it; but it was interesting to read about it.
TBF, the "developer cycles matter more" crowd still has to eat crow sometimes; it's just that it's usually not all the way down to assembly.
Appended to the end of comments you post. Max: 120 chars.
(Score: 2) by DannyB on Wednesday March 13, @06:57PM
On the "developer cycles" thing, that is an economic consideration.
Sometimes the actual pure raw speed is the overriding economic consideration regardless of development cost. That can be a valid scenario. It just isn't the most common scenario these days.
People who think Republicans wouldn't dare destroy Social Security or Medicare should ask women about Roe v Wade.
(Score: 3, Informative) by gnuman on Thursday March 14, @09:57AM (2 children)
Assemblers are still nice to use for microcontrollers. Heck, these are almost necessary if you need to have exact timing on these. C is also very useful on these systems if they have more resources, like the RPi Pico. But seriously, why write in C on something that has 2kB program and 128 bytes of RAM like on ATTINY202?
Assembly is actually simpler to understand in some of these situations. YMMV
(Score: 2, Funny) by Anonymous Coward on Thursday March 14, @12:07PM
I find writing SIMD code to be easier to read, write, and think about in assembly than intrinsics. For the rare functions involving many casts or complex pointer stuff I'll often write in assembly for ease of thought and then translate into high level for portability. Sometimes I find it easier to factor out and simplify code compared to higher level languages due to less visual clutter. I also get much more joy out of it.
(Score: 2) by JoeMerchant on Tuesday March 19, @09:00PM
I'm not sure why anybody bothers with processors smaller than the RP2040 that Raspberry Pi Pico is based on... I mean, sure, if you've got a production volume in the tens of thousands or more, then the chip cost _might_ come into play, but below that volume the additional engineering hours needed to work with the cheaper per copy chips is usually going to swamp any per-unit cost advantages. Not only are the smaller chips more labor intense to shoehorn functionality into, but if you're squeezing the last penny out of the CPU you also usually end up having to maintain different toolchains for every project. In the end, the CPU is usually one of the cheaper components in the system, anyway.
Back in the 1990s, we settled on the Moto 6811 family as our "baseline" for projects. Usually had the bigger chips on hand to do onesey twosey projects, and if we were going to make a bunch of something we'd try to make it fit in a cheaper chip like the D3, but the toolchain was the same for all. As I was job searching in the early 2000s, I discovered that Caterpillar was using 6811s to control their big excavators, graders, front end loaders, etc.
🌻🌻 [google.com]
(Score: 2) by JoeMerchant on Tuesday March 19, @08:53PM
>I thought I could do anything in BASIC
You thought correctly: given enough time.
After a couple of years programming in BASIC I started to hand-compile 6502 assembly code into strings that BASIC would call to speed up bottlenecks. That's sort of still programming in BASIC...
🌻🌻 [google.com]
(Score: 0) by Anonymous Coward on Sunday March 24, @02:37AM
I've heard similar things said about Lisp: https://en.wikipedia.org/wiki/Greenspun%27s_tenth_rule [wikipedia.org]
While in theory you could do anything in BASIC, it's easier to do some things in Lisp: https://www.cs.ox.ac.uk/activities/ieg/e-library/sources/georgia.pdf [ox.ac.uk]
Of course it is also easier to do some other things in Perl/Javascript/etc.
(Score: 2) by turgid on Sunday March 17, @03:49PM (1 child)
C has all sorts of problems. I say this as an enthusiast who loves C but it has all sorts of rough edges. I use it in preference to C++ whenever I can. I think it's a real shame the Pascal family of languages didn't become more popular. Borland's Delphi language was very nice and elegant and had great potential. If Delphi was Betamax, C++ was VHS, and look what happened. C++ should have been a dead end, a failed experiment to extend C but it unfortunately gained traction in the market and the dead horse has been flogged ever since.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 1, Funny) by Anonymous Coward on Friday March 22, @06:14PM
Yes, but that is part of the challenge and glory of C. While C is indeed like Highlander, it is also like Spider-Man -- with great power comes great responsibility.
Most "modern" computer languages have been invented not due to the need to handle increased computing capabilities, but because more and more "programmers" are stupid.
(Score: 3, Insightful) by rufty on Tuesday March 12, @11:55PM (3 children)
Since "We have too many already" is in the lead right now, we need another poll: which languages to axe. My vote? Modula 2
(Score: 5, Funny) by Rich on Wednesday March 13, @12:48AM (1 child)
Nooo! That was the last beacon of sanity in the language zoo. I propose either C++, where the reasonably sane appearing community needs help to get rid of its insane language, or Rust, where the reasonably sane appearing language needs help to get rid of its insane community.
(Score: 2, Disagree) by Anonymous Coward on Wednesday March 13, @01:53AM
Rust and C++ are practically visually identical, just a matter of more colons or more angle brackets.
(Score: 2) by VLM on Friday March 15, @02:05PM
I've seen this discussion play out a million times and it always devolves to:
1) Work or School forced us to use a poorly chosen language so we hate it by association. See "C++"
2) It's old enough that the "good idea fairy" has hopelessly trashed the language and/or the libraries and/or the general culture and design of that language. If you learn enough of the huge ecosystem you'll learn the hard way there is no intelligent design behind it, there's just a pile of "it wasn't screwed up enough to motivate someone to successfully fix it". See "R", what an absolute pile of fertilizer. Perl doesn't have to be fertilizer, none of mine was, but "in the old days" a lot of fertilizer factories used Perl so now "everyone has to hate Perl" in 2024. Those same fertilizer factories use Python now, watch this space in a couple years LOL...
3) It's not a problem with the language, but the docs or development experience overall was so awful that everyone considers the language "awful". How about BASIC? If you mean MS-BASIC on a home computer in the 80s where the only error message was "?SN ERROR" then you'll hate BASIC. If you used Microware's BASIC09 maybe you'd have a happier memory of BASIC.
4) It's essentially a punishment or make-work language. See Java. Assembly language on "extremely limited" hardware like the most ancient PICs.
(Score: 2, Disagree) by GlennC on Wednesday March 13, @07:54PM (1 child)
As long as "Not Invented Here" continues to be a thing, new languages will continue to be created.
Sorry folks...the world is bigger and more varied than you want it to be. Deal with it.
(Score: 2) by turgid on Sunday March 17, @02:56PM
Sometimes it's just fun. Sometimes someone says, "What if..."
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 4, Funny) by PiMuNu on Thursday March 14, @06:06AM (6 children)
C++ is a good example of why we need maw languages.
C++ was probably a good language. Then every 3 years it bloats. At some point it reaches an inflection point where no one can understand it and at that point we move to a new language. Enshittification for code.
(Score: 4, Informative) by janrinok on Thursday March 14, @07:55AM (5 children)
What you have written is correct, but nobody is forced to use the new enhancements - or even to understand them. In the vast majority of cases (all?) the 'original' C++ still works.
If you have a specific problem that C++ can't handle for you then maybe you are justified at looking for a new feature that does what you ask and, if you find it, then presumably it you are prepared to invest the time to learn how to use that feature properly.
I must admit that I do not use C++ very often nowadays. Python is my preferred language, and D if I actually need the extra speed. Python still processes data at a faster rate than my brain can assimilate it so there is usually no need to optimise for speed, but rather for optimise for the least amount of effort and time spent producing software to do whatever I want it to do.
D suffers from a lack of libraries, but a recent initiative to change this appeared in January of this year. I suppose I should look at Rust again.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 3, Insightful) by turgid on Sunday March 17, @02:58PM
Python is a good language, but it's being used these days for anything and everything where other solutions would be more appropriate. I've seen relatively small embedded systems with strict power and performance requirements hosting Python code that really shouldn't be there. It's exactly the sort of stuff that should be written in plain C.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 3, Interesting) by JoeMerchant on Tuesday March 19, @09:04PM (3 children)
Once in awhile those new enhancements in C++ make a lot of sense and feel helpful, but I'll agree: 90% of them are best ignored.
> Python still processes data at a faster rate than my brain can assimilate
My main complaint about Python is that I usually find that it has a bottleneck somewhere - bit fiddling an image was the most recent - I'll be going along in Python and it's 98-99% as fast as C++, then you hit that bottleneck and it's 1000x slower... And the Pythonish answer to that is to create a module in C or Fortran or whatever to handle the bottleneck, at which point I feel like: well, why am I bothering with this dependency hell if I _also_ have to throw in other languages to get a complete solution?
🌻🌻 [google.com]
(Score: 1) by shrewdsheep on Thursday March 21, @08:16AM (2 children)
And that's exactly it. As long as scripting languages simply wrap around C, they are as fast. As soon as the algorithms are implemented natively, they suck. Interpreted languages should introduce optional strong typing to allow compilation. Old example: https://lush.sourceforge.net/ [sourceforge.net] recent: https://raku.org/ [raku.org]
(Score: 0) by Anonymous Coward on Thursday March 21, @09:46AM
Strong typing isn't a necessity to have a compiled implementation language. You mentioned C as a counterexample and C both weakly typed and has many compiled implementations. You can have any combination of strong/weak typed languages and compiled/interpreted implementations.
(Score: 3, Informative) by JoeMerchant on Thursday March 21, @12:09PM
The Qt API implements weak typing through QVariant over C++... As well as QJson and probably others I don't use much...
🌻🌻 [google.com]
(Score: 2) by ese002 on Friday March 15, @01:29AM
As long as programming is a thing and, problem spaces change, and existing languages imperfectly address the problem space there will always be a need for new languages. When problem spaces stagnate, a point can be reached where the cost of moving to a new and "better" language outweighs the benefits. However, we have not reached the point where all programming spaces are stable. With AI, we may reach the point where programming is no longer a thing for humans first.
(Score: 2) by VLM on Friday March 15, @01:43PM (1 child)
I claim "no" because of the cyclical ever repeating nature of history.
All the shittiest MSBASIC programmers created shit code so they gave MSBASIC a bad rep. Ah this new VB6 has no bad rep. Until all the now-unemployed MSBASIC programmers moved to VB6 and shit it all up as they do.
All the shittiest VB6 programmers created shit code so they gave VB6 a bad rep. Ah this new Perl has no bad rep. Until all the now-unemployed VB6 programmers moved to Perl and shit it all up as they do.
All the shittiest Perl programmers created shit code so they gave Perl a bad rep. Ah this new Python has no bad rep. Until all the now-unemployed Perl programmers moved to Python and are in the process of shitting it all up as they do.
Java and JS and a couple of others belong in the above list somewheres but you roughly get the right idea.
There will be infinite new rounds of this forever. Will it be Go? Rust? Kotlin? Webassembly (LOL)? Probably not Haskell or a LISP, but, hypothetically, it could happen...
(Score: 0) by Anonymous Coward on Saturday March 23, @05:52PM
You forgot php. How did you manage that? I've been trying to forget it for years!
(Score: -1, Offtopic) by Anonymous Coward on Saturday March 16, @06:41AM (58 children)
We ACs who have been banned from commenting on SoylentNews need to make it clear that once you have eliminated venues for us to express our opinions, as members of the SoylentNews Community, necessarily have recourse to the one left open to us. It is not really appropriate to post off topic comments in members journals, unless they invite it. But this is the commons, the one forum left. I only fear that certain forces will seek to quash it. And when they do, SoylentNews will be Dead.
(Score: 2) by janrinok on Saturday March 16, @07:42AM (42 children)
Only one account is permanently banned. There are 2 accounts with a temporary ban, which is usually for non-compliance with the site policy and rules.
Commenting in journals is controlled by the owner of the journal. The staff have no control over who can or cannot comment in a journal.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Saturday March 16, @04:18PM (40 children)
Sorry to ruffle your feathers, but the deletion of comments you deemed unacceptable violates your interpretation of facts.
(Score: -1, Troll) by Anonymous Coward on Saturday March 16, @06:10PM (8 children)
Truth is a troll? For shame lamers.
(Score: 2) by janrinok on Saturday March 16, @07:15PM (3 children)
There have been 19 copies of the following spam, with minor variations, posted here. Which of them do you think is relevant to the topic under discussion?:
This is a single example of the spam that is being dumped in people's journals and in the poll. There have been over 500 spam posts in less than a month. They have all been removed.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Saturday March 16, @08:55PM
I was not arguing the merits, but given the previous mixup and reasoning "sounds like aristarchus" for censorship of non-spam I felt compelled to hold you accountable to the facts. Modding down such facts, when done in a polite fashion, shows your inability to accept feedback without retaliation. Could have been someone else, but then they should be mod banned for abusive moderations. I had no cause to complain recently until now, stop assuming the worst and making up intentions of others.
(Score: 0) by Anonymous Coward on Monday March 18, @12:37AM (2 children)
Truth-trolling! Ah, back in the day! "Give'm hell, Harry!" To which President Truman replied, about the Republicans, "I just tell the truth, and they think it's hell." Somethings never really change.
(Score: 1) by pTamok on Tuesday March 19, @03:27PM (1 child)
Jus' remember, "Reality has a well-known liberal bias."
(Score: 1) by pTamok on Tuesday March 19, @03:35PM
...and I should have said, we need at least one more programming language.
Why?
Because all our attempts, so far, to produce a language in which us crazily inconsistent, irrational humans can express our requirements has utterly failed. We need a 'Do what I mean, not what I say' language, which includes a pre-processor that converts the discovered intent into what we need, and not necessarily what we want. The language preprocessor will necessarily be more intelligent than we are, able to answer philosophical questions that several hundred years of navel inspection by clever minds have so far failed at.
That is why we need one more programming language. Rust doesn't meet the objectives.
(Score: 1, Insightful) by Anonymous Coward on Saturday March 16, @09:30PM (30 children)
So once in a while you wake up and the music is playing in your head,
and you find you have an idea that could fit neatly into a journal
that might spark some interesting discussion amongst thinking adults,...
so you post it.
And for a while it is all good
and you learn some things
and you share some things
and some get upmodded
and some get downmodded
but it is all good fun
until some guy shows up with a chip on his shoulder
and an agenda to accomplish
as if he could save anyone in the world from extremism and wrong-think
and oppression
and as if we were all children in need of a nanny
And suddenly there is an ascii rainbow nazi parade being posted to your journal
and people are obscenely employing words found in Firesign Theater's song What Makes America Great [lyricsbox.com]
and the well of words gets poisoned
and you wind up withdrawing your journals simply to avoid any guilt by association
and a journaler's freedom of speech and artistic prose is shat upon.
So when an admin makes an effort
to rescue some journal from descent into madness
one can't really complain
or feel terribly sorry that some bad actor felt mistreated.
It is the price of keeping nice things nice.
Still, I get it... it just ain't fair.
-nostyle
--
-The Beatles, I’ve Got A Feeling (Take 10)
(Score: -1, Troll) by Anonymous Coward on Saturday March 16, @10:21PM (28 children)
https://soylentnews.org/comments.pl?noupdate=1&sid=60163&page=1&cid=1349112#commentwrap [soylentnews.org]
I do not think holding staff accountable to the truth is a big request, the site has a history of abusive admins and false accusations. Where did you see a request to not remove nazi garbage? Also, if you have issues with AC spam then simply disallow AC comments in your journal.
(Score: 0) by Anonymous Coward on Saturday March 16, @11:30PM (4 children)
I was never into helicopter parenting...but I did have to stop the car from time to time. The kids "got" it eventually.
Anyway, all the best comments are from AC.
(Score: 3, Insightful) by janrinok on Sunday March 17, @12:52AM (3 children)
There have have been at least 2 attempts to dox somebody so far this year, and several phishing attempts to confirm information that someone believes he already knows. It is not a problem that has gone away. Allowing ACs to spam (and I have no wish to prevent them from having a discussion) means that it is easier to slip a doxxing attempt through unnoticed.
If the journal owner allows AC posting, is he also responsible for any doxxing that happens in his journal? Or, does preventing doxxing remain the responsibility of the administrators? I suggest that you continue as you already do and allow whomever you wish into your journal, and the administrators will continue to look after the other issues. As the spamming and the doxxing is often from the same source then I believe our current approach is a reasonable one.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by janrinok on Sunday March 17, @12:31AM (22 children)
We have had this extended discussion in another place (where it was also off-topic!). I have responded to all of your questions. I have told what you need to do if you want something to be changed. You do not seem interested in actually making any changes, yet here you are creating the same discussion again (and where it is also off-topic).
It is not unreasonable therefore to assume that the actual purpose of restarting the conversation is simply to disrupt this poll. It is not here as a general chat area for ACs. Your repetitious comments should they continue could quite reasonably be moderated as Spam.
Furthermore, as you seem determined to induce me into another long and pointless discussion for which I have already provided the answers, it appears that you are trying to troll me. I will therefore moderate it as I see it.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Sunday March 17, @01:25AM
Ha! I made a single comment that was pointing out a fact. The comment you replied to was meant for nostyle's comment. I don't enjoy these long threads, but you frequently keep posting BS about me and try to tie me to someone else. You could just let my factual statements stand instead of trying to deny things. Also, mod abuse without being a proper offtopic sometimes gets further comaints.
Again, you could always stop abuse, at least this time it is just lame troll mods. I have never argued that true spam should not be removed, only when you swept me up in it. Want less lengthy threads? Ignore my comments, or reply to the comments themselves instead of trying to smear me as the actual spammer.
Simple stuff.
(Score: -1, Troll) by Anonymous Coward on Monday March 18, @03:56AM (1 child)
So as usual you engage an offtopic thread, I chime in to correct a factual error, and you blame me for it all. You clearly never learned humility, maybe next life you will do better. That is not a threat btw oh sensitive little abuser.
This comment brought to you by your own petty troll mods when you couldn't just admit to censoring spam that caught up someone other than your target. I fully expect you to cement this reality with further abuse, unless you want to make things even and troll mod your own spammy replies to the real spammer. Without your incomplete reply I would not have entered the discussion and thus would not have received your continued abuse.
(Score: -1, Troll) by Anonymous Coward on Monday March 18, @08:29PM (1 child)
I see you have yet to troll mod the janrinok comments in this heinous offtopic thread you got rolling.
(Score: 0) by Anonymous Coward on Tuesday March 19, @05:57PM
Another troll mod, I guess abuse is your natural state.
(Score: -1, Troll) by Anonymous Coward on Thursday March 21, @06:15PM (14 children)
Perhaps it would be useful to allow AC comments on the Meta posts on the front page, since ACs are a large and traditional constituency on SoylentNews. Or have we all be painted with a broad brush and silenced? A bit more transparency and inclusiveness is a good idea.
(Score: 2) by janrinok on Thursday March 21, @06:47PM (13 children)
No. You are NOT a large group when compared to our full membership Nor are you silenced. You have chosen to be anonymous. It has not been forced upon you.
.
We've tried it and it doesn't work. Get an account and your problem is solved. You can still post anonymously if you have an account and there are no restrictions.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Friday March 22, @12:00AM (12 children)
I would advise against this. It has been repeatedly demonstrated that admin on this site have no qualms about doxxing regular users who post anonymously. This is not a solution. And the numbers do not add up and appear to be a fabrication.
(Score: 2) by janrinok on Friday March 22, @02:31AM (11 children)
If you can guarantee that every AC will play by the rules then we could do this. But that is not what happens, is it?
There is only one Anonymous Coward account. We either permit all ACs to rejoin the main stories or we prevent all ACs from doing so. Your AC account is no different from any other. If you want to be treated differently the then you have to have something unique about you. That something is an account.
What happens is that some ACs will use the opportunity to use the front page access to verbally attack others, to post spam and disrupt normal discussions, and to complain about off-topic issues that are nothing to do with the Meta content.
I haven't quoted any numbers. Which numbers do you think are fabricated?
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Friday March 22, @08:28AM (10 children)
Maybe if ACs could comment on the Meta stories, their comments would have something relevent, and not need to be posted in the poll? Seems logical to allow them. Are we afraid of demented persons like apk, MikeUSA, or Donald Trump insulting us? I think I could ignore that.
(Score: 2) by janrinok on Friday March 22, @09:48AM (8 children)
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Saturday March 23, @03:49PM (7 children)
New AC here, you glossed over the fact that logged in user's "AC" comments are tied to their username and thus not AC at all since staff have previously unmasked such comments. You linked to a policy document once which stated that info, but I am not seeing a current link to it. Important bit for those trusting words to have reliable meaning.
PS: please for the love of Alwhesus do not mod informative comments as troll
(Score: 2) by janrinok on Saturday March 23, @03:59PM (6 children)
No they are NOT. We have explained this many times. It is even documented somewhere in the Wiki. If you cannot work out how to achieve it try searching for the information. So your comment isn't informative - it is just plain wrong.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Offtopic) by Anonymous Coward on Saturday March 23, @06:57PM (5 children)
Are you confused, or deliberately lying? You have confirmed the fact yourself many times, it just seemed your recent comment was skirting around it.
So if you janrinok make a comment and use the post as AC option then the comment is NOT tied in the database to the janrinok ID? I recall that it is tied to the user ID with explanations of tracking karma. That and staff have exposed what user account was behind an AC comment in the past.
So I'm giving you the benefit of the doubt that you merely misinterpreted the context. Otherwise you are a big huge liar. Or you have assumed I meant "publicly tied to user name" which is being willfully obtuse especially if you do not explain the possible confusion.
So, are you a liar? Or assuming jerk face that can't handle even the slightest bit of correction or criticism? I certainly wasn't trying to attack you or troll the discussion, so at minimum you are abusing the mod system AGAIN!
Or are you truly stating that you cannot tell a non-user AC post from a user AC post as an admin?
(Score: 2) by janrinok on Saturday March 23, @07:58PM (4 children)
Correct, it is not connected to my account ID. I might compromise who I am in other ways (the penny will drop for you eventually....), but there is no connection to my account.
Then you recall wrongly. ACs do not have karma. There is no 'tracking' of karma. There cannot be - every AC is using the same account. Postings that I make as AC do not affect my account karma. That is one reason why logged-in users like the feature. They can say something unpopular but cannot be moderated down for saying it in a way that will affect their account.
AC posts are not linked to an account - privately or publicly by the software. You have access to the code. You can sort it out for yourself. I am not going to discuss security weaknesses here in public. If you are interested in our security join the staff. We can keep you busy.
The topic for discussion is a Poll. It is not a general place to be used by ACs for discussing any matter that they wish. I have provided a list of journals on IRC that you can use for any discussion that you want to have. However, as you (AC) keep repeating the same messages to me then at best you are trolling me, or you are spamming the Poll. Get it into your head that the Poll is NOT a general discussion journal. If I mark your posts as spam there are other repercussions for you - and your comments might end up being deleted.
Only if you compromise yourself. You see, you have all been worrying about something that does not exist. There is NO tracking of AC accounts or their comments in Rehash. Look at the code. Show me where it does it.
If you want to comment on the front page stories - which includes the Poll - then get yourself an account. It is quite simple - you just have to prove that you are not a bot. You do not get all privileges regarding commenting until you reach a certain karma level. Getting an account and then just posting as AC will not get you that karma because ACs do not get karma.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Saturday March 23, @09:18PM (3 children)
You recently linked to a site policy document not hosted on soylentnews.org that said the opposite, that registered user AC comments are tied to that user ID. We have discussed this many times and you eventually came clean on how it works. I find it odd you are reversing course now, plus we have the history of admin abuse where identities were exposed and alluded to.
Oh well, guess you are as evil as suspected but for what purpose I do not know.
(Score: 2) by janrinok on Saturday March 23, @09:45PM (2 children)
The policy document is hosted on soylentnew's wiki: https://wiki.staging.soylentnews.org/wiki/SoylentNews_Site_Policy [soylentnews.org] You can access the wiki from a link on the lefthand side of the front page unsurprisingly marked as 'Wiki' in the SoylentNews panel.
If someone first posts a comment under their username and then posts a comment as AC - without changing their IP address - then it is obvious that it is coming from the same connection. Most people don't give a damn about this. If you don't understand how it works there is little I can do to help you.
If an AC posts 2 comments from the same connection then I can see that they are from the same person. It does not tell me who that person is. But usually they compromise themselves. Like you started this discussion by saying 'New AC here' but now you have linked yourself to our previous conversations. You have compromised yourself. You are just not smart enough to realise it. You are not a new AC - you are the same AC that has been trolling for months now.
If you want to stay anonymous then change your IP address after EVERY comment and do not refer to things that you have written previously. There you go - anonymity guaranteed. Don't keep repeating the same complaint. It is also clear that this is spam - repetitious and off-topic.
It isn't software that is doing this - it is simply that people are more observant than you imagine that they are.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: -1, Troll) by Anonymous Coward on Saturday March 23, @09:58PM
Lol ok buddy, keep lying to everyone
(Score: -1, Troll) by Anonymous Coward on Saturday March 23, @10:09PM
PS: I am not trying to stay anonymous, this "trolling" or fact checking started when you began calling me a banned user and allowing moderation abuse. I have been upfront many times about my own commenting behavior, yet you persist in making assumptions and now plain lies. I will continue to call out your lies and abuse, feel free to stop any time you'd like. As for this latest lie, gaslighting seems to be your specialty and site documentation has been updated, so looks like you are dedicated to it. Insulting people for what you think they are doing, as opposed to what is actually happening, says a lot about you. And the lies.
(Score: 2) by janrinok on Friday March 22, @11:50AM
I can give you regular updates to new journals on IRC if you wish.
They are all commercial accounts which have been disabled but their journals will still accept comments from ACs. Somehow, though, I don't think you will use them...
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 2) by janrinok on Saturday March 16, @08:00AM (1 child)
I suspect that you have a temporary ban. If I am correct then the ban has been extended because of your own actions. You were warned several times that this would happen.
Play stupid games, win stupid prizes.
I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(Score: 0) by Anonymous Coward on Saturday March 16, @04:21PM (10 children)
There are no commons, you can petition the government to provide such a space or regulate private entities how you would like. Good luck, don't think many will support your idea.
What did you get a temp ban for?
(Score: -1, Spam) by Anonymous Coward on Friday March 22, @06:26PM
SoylentNews is already dead. It just hasn't noticed yet, and thus is still kind of flopping around.
Before the ban on ACs, a story would get dozens and sometimes hundreds of replies. Sure, a lot of it was everyone yelling at each other about being stinky poopyheads, but it was something. Now, most postings get three or four replies. That's not a site that's alive, that's coasting on user nostalgia for what once was. The AC ban did indeed get rid of the stupid trolling ACs, but at the cost of making the site boring.
Maybe once all the behind the scenes legal stuff about transferring ownership gets worked out, some new life can be breathed back into the shambling corpse.
(Score: 2) by turgid on Sunday March 17, @02:52PM (6 children)
Will they ever stop adding features to C++? Will someone some day finally say, "Stick a fork in it, it's done?" Pretty please?
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 1, Touché) by Anonymous Coward on Sunday March 17, @10:53PM (4 children)
No. According to the C++ philosophy, it will never be complete because the problems and the field itself are always going to change. Therefore, the language itself must change with the times. The programmer, on the other hand, is perfectly free to choose a specific standard, library set, or feature set applicable to their problem and ignore the rest. The people in charge have said that if you don't then you are the one using it wrong.
(Score: 3, Insightful) by pTamok on Tuesday March 19, @03:38PM (3 children)
It will never be complete, because there will always be questions that can be asked in C++ that cannot be answered in C++. (Gödel)
(Score: 0) by Anonymous Coward on Wednesday March 20, @02:24AM (2 children)
Neither of Gödel's incompleteness theorems tell you anything about C++ due to a whole host of issues including Rice's theorem, the underlying logic being inconsistent, the lack of effective axiomatization, and a bunch of other even more technical concerns.
(Score: 2, Interesting) by pTamok on Wednesday March 20, @08:38PM (1 child)
Well, clearly, following Rice, it is possible to write programs in C++ for which the end-state is undecidable. You might like to reflect on the connection between the halting problem and Gödel's famous incompleteness theorems.
In fact, following Rice, it is clearly not possible for any non-trivial computer programming language to produce only decidable programs, and therefore no programming language is 'complete' (all possible outputs are decidable).
We then have to fall back on arguing over whether any particular programming language is suitable for our purposes/use-case - which can be an undecidable matter of taste.
(Score: 0) by Anonymous Coward on Wednesday March 20, @11:01PM
Complete isn't the same thing as decidable. I will grant that they are fundamentally related, but that doesn't meant that you can put a biconditional implication between them in all cases. Regardless, any proof that C++ were incomplete wouldn't come from the Gödel's incompleteness theorems. As I already pointed out, they don't apply to C++ because, at a minimum, the formal system underlying C++ is neither consistent nor effectively axiomatized.
(Score: 3, Interesting) by bart9h on Tuesday March 19, @01:55AM
No, but I hope cppfront will get traction.
(Score: 2) by Mojibake Tengu on Wednesday March 20, @11:31AM
Best programming languages are easy to determine: If you want to write a new programming language, what existing language do you use as a base?
Rust programming language offends both my Intelligence and my Spirit.
(Score: 2) by mendax on Wednesday March 20, @11:33PM
I don't exactly know how to code in binary, I have learned four assembly languages in my time: CDC 6600 (Seymour Cray's first supercomputer), PDP-11, TI-9900, and 6502. I tried learning 32-bit x86 assembly but it seemed bizarre to me. Given all the additional crazy special-purpose instructions that family of processors has now it would be insane to try to learn it. Now that I have an M1 iMac it might be fun to learn ARM64, especially, like, CDC 6600, it's RISC and should make a lot more sense. Of course, having said that, when I first started writing assembly language programs, compilers generally weren't very good at optimization. They've come a long way since then and a modern C compiler, for example, can generate some pretty tight code. But who worries about optimization these days, especially when a CPU can run billions of instructions per second, when one is not compiling an OS kernel which, by necessity, needs to be fast and tight?
It's really quite a simple choice: Life, Death, or Los Angeles.