TIOBE's Surprisingly Popular Programming Languages: R, Go, Perl, Scratch, Rust, and Visual Basic 6 (techrepublic.com) 101
The R programming language is experiencing a surge in popularity "in the slipstream of Python," according to this month's TIOBE index, leaping into the top ten.
"For historical context, we wrote of R's spot in TIOBE nearly two years ago, and it had just made the leap from #50 to #39," writes programming columnist Mike Melanson.
ZDNet writes: In May, when R crashed out of the top 20 for the first time in three years, Tiobe speculated that the language could be a victim of consolidation in statistical programming, with more developers in the field gravitating towards Python.
But there's been a lot of motion since then, Tech Republic reports: R rose one space to eighth place in July, but its comparison to 2019 is where the real surprise lies: It was in 20th place at the same time last year. TIOBE CEO Paul Jansen cites two reasons why R may be increasing in popularity:
- Universities and research institutes have moved away from commercial statistical languages like SAS and Stata in favor of open source languages Python and R.
- The increase in analytics being used to search for a COVID-19 vaccine....
The largest gainers in popularity between July 2019 and July 2020 are Go, which jumped from 16th to 12th place, Perl, jumping from No. 19 to No. 14, Scratch, jumping from No. 30 to No. 17, Rust, which moved from No. 33 to No. 18, and PL/SQL, which moved from No. 23 to No. 19.
Ruby fell the most, moving from 11th place to 16th, while SQL, MATLAB, and Assembly Language also slipped down the list.
ZDNet adds that "Besides R's upwards shift, Tiobe's July index doesn't show much movement in the popularity of the top languages. The top 10 in descending order are C, Java, Python, C++, C#, Visual Basic, JavaScript, R, PHP and Swift."
Visual Studio magazine argues that the biggest surprise may be that the 29-year-old language classic Visual Basic is still in the top 20 — since its last stable release was 22 years ago, and by 2008 it was finally retired by Microsoft. "VB6 just refuses to go away, achieving cult-like status among a group of hard-core supporters."
"For historical context, we wrote of R's spot in TIOBE nearly two years ago, and it had just made the leap from #50 to #39," writes programming columnist Mike Melanson.
ZDNet writes: In May, when R crashed out of the top 20 for the first time in three years, Tiobe speculated that the language could be a victim of consolidation in statistical programming, with more developers in the field gravitating towards Python.
But there's been a lot of motion since then, Tech Republic reports: R rose one space to eighth place in July, but its comparison to 2019 is where the real surprise lies: It was in 20th place at the same time last year. TIOBE CEO Paul Jansen cites two reasons why R may be increasing in popularity:
- Universities and research institutes have moved away from commercial statistical languages like SAS and Stata in favor of open source languages Python and R.
- The increase in analytics being used to search for a COVID-19 vaccine....
The largest gainers in popularity between July 2019 and July 2020 are Go, which jumped from 16th to 12th place, Perl, jumping from No. 19 to No. 14, Scratch, jumping from No. 30 to No. 17, Rust, which moved from No. 33 to No. 18, and PL/SQL, which moved from No. 23 to No. 19.
Ruby fell the most, moving from 11th place to 16th, while SQL, MATLAB, and Assembly Language also slipped down the list.
ZDNet adds that "Besides R's upwards shift, Tiobe's July index doesn't show much movement in the popularity of the top languages. The top 10 in descending order are C, Java, Python, C++, C#, Visual Basic, JavaScript, R, PHP and Swift."
Visual Studio magazine argues that the biggest surprise may be that the 29-year-old language classic Visual Basic is still in the top 20 — since its last stable release was 22 years ago, and by 2008 it was finally retired by Microsoft. "VB6 just refuses to go away, achieving cult-like status among a group of hard-core supporters."
Imagine my complete lack of surprise (Score:2)
You're doing it wrong (Score:2)
Instead of
You should have wrote
"You keep using that word. I do not think it means what you think it means." — Inigo Montoya, The Princess Bride
Re: (Score:2)
Thank you.
Re: You're doing it wrong (Score:1)
Re: (Score:2)
Not a bot. If people don't point out the mistakes I make when I write in english, I'll keep on making the same mistakes in the future.
Re: (Score:2)
Not a bot. If people don't point out the mistakes I make when I write in english, I'll keep on making the same mistakes in the future.
Pretty impressive bot ... so contextually accurate.
Re: (Score:3)
Thank you kind sir, but still not a bot. Maybe you would like to buy some contextually accurate [amazon.com]?
Re: (Score:2)
This was *beautiful*.
Perl surprises people (Score:5, Interesting)
Perl gets a bad rap. In part this is because people don't learn it anymore. they learn python which for structured programing applicaitons is better. they giggle when they see perls and all it's sigils. And so they don't bother to look into it. Perl requires druidic wizard skills but it rewards them with power to do amazing things with the flick of a wand rather than writing an elaborate program.
In the early days of bioscience with the concurrence of both federated web processing of forms and the need for heavy duty string processing (genomes) perl was tailor made for the job. And it also was a nice glue language for reformatting one programs output into another's input. There wasn't a better language and at the time no one had heard of python.
Python seemed to first gain traction on Windows platforms I think because Perl wasn't really available on windows, you could use cygwin, but even when you got it working perl's model of the file system was kinda unix centric it seemed.
But these days I find a lot of people rediscovering perl and simply being shocked at what a efficient language it is for getting jobs done than involve gluing programs together or parsing flat files.
The way people back intothis is that at some point after graduating from running apps, learning to install an build codes they need for their jobs they start trying to put these chinks of other-people-software together and into a pipeline and end up finding they need to learn Bash (or similar).
They get so proud of themselves for learning Bash (hey I know Linux!). And then you introduce them to Perl. And their jaw drops and they discover how their painful experiences trying to do anything no trivial in Bash, calling Awk a bazillion times, was such a waste of time.
Perl is a woodchipper for text. And you can do so much on one line.
It's best uses are for system utilities you need only write once for some ad hoc job you want to automate, or a quick and dirty on-time reformat, like say slurping some thing off some website. If you need to write programs that require correct design and complexity and encapuslation then python is a better choice.
But perl is quite amazing, and I'd say it's absolutely true that no one should bother learning to program in BASH ever, just use perl. Perl is also extremely fast for opening reading and parsing flies, and in my experience is generally faster than python for this by very large measures. Python has recently caught up but only through the use of specialized libraries (that is, C-code hiding underneath) rather than Perl where the speed is in native commands.
O'reily nutshell (Score:2)
For me a anecdotal indicator of perl's amazing powers in the native language itself (without libraries) is the thickness of the O'reilty nutshell book. It's the thinest one on my shelf. The result is that even though perl is often lambasted for being a collage of disjoint tools it's actually not that. It's actually very few but extremely powerful ones that replace the libraries other languages relegate things to. THis is why the book is so short. It's shorter than the C++ o'reily nutshell. And the pyth
Re: (Score:2)
Amazon claims that the most recent Perl in a Nutshell book has 762 pages, while the most recent Python in a Nutshell book has 772 pages. So while it's technically thicker, I'm not sure how one could distinguish that.
Having said that, I can understand how Perl could be misunderstood.
and I'd say it's absolutely true that no one should bother learning to program in BASH ever, just use perl.
I'd probably rather use D over both of them for that, what with rdmd [dlang.org] and such.
Off by an order of magnitude! 67 pages (Score:2)
My perl 5 nutshell book is 67 pages.
You are looking at the "New" perl nutshell. Which sadly isn't the same. and probably is an indicator of how people are making perl too hard.
Re: (Score:2)
Re: (Score:2)
dude I have the second edition. it's in my hand. it's 67 pages. You must be using the internet wrong :-)
Re: (Score:2)
okay... my bad. not nutshell. should have said O'Reily "pocket reference".
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
perl -e '$_=(join "",reverse sort <) ; s/F/FF/g; warn $_ ' /var/log/system.log
Re: (Score:3)
sort entire file, reverse the order, concatenate the list and change all the F's to FF, send it to the standard error
Well, the same can apparently be done in
rdmd --eval='stderr.writeln(stdin.byLineCopy.array.sort.reverse.join("\n").replace("F","FF"));'
Maybe it's slightly longer, but personally I find it more comprehensible. I'm not even a D programmer and it only took me like five minutes or so to find out in the documentation how to do all this (plus there's a decent chance it could be made shorter than this even using just the standard features). Not quite sure how long I'd spend with Perl documentation regarding how
Re: (Score:2)
No one asks, "why would you want to do all of that on one line?"
Readability is the only metric that matters five years after the code is written.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Why does this bleed past the boundaries of the container?
rdmd --eval='stderr.writeln(stdin.byLineCopy.array.sort.reverse.join("\n").replace("F","FF"));'
Re: (Score:2)
Re: O'reily nutshell (Score:1)
Re: (Score:2)
Most people who use Python really don't have anything to compare it too. I have about a dozen languages in both major paradigms under my belt. Give me Perl and Groovy before Python. Esp. Groovy.
Re: (Score:2)
I know little about groovy. When it came out I read it let you write scripts then later add types and compile it. Which sort of sounds like a holy grail of rapid development to stable product. But I've never looked at it.
Tell me more. what do you like?
Re: (Score:3)
Re: (Score:2)
Strange, I've seen more unreadable python than perl.
Probably because most of the perl I've seen has been short scripts written by people who don't think everything needs to happen in 12 characters. Most of the python I've seen has been larger programs where python should never have been a choice.
The only short program that comes to mind was a maze generator someone wrote in python. It was broken, which is why they asked me about it. It was supposed to produce a binary tree maze. (For those that don't know
Write once read never == bad investment (Score:2)
Perl gets a bad rap. In part this is because people don't learn it anymore. they learn python which for structured programing applicaitons is better. they giggle when they see perls and all it's sigils. And so they don't bother to look into it. Perl requires druidic wizard skills but it rewards them with power to do amazing things with the flick of a wand rather than writing an elaborate program.
You just stated the fatal flaw. It doesn't matter what you can do with it. You may be awesome. It may be the holy grail panacea in your hands, but think about it from the perspective of the guy writing checks. You would never allow a plumber to use custom fittings he designed himself if it mean no plumber but him could ever repair or upgrade it. No one cares how great you are with it.
The people paying for my time as a professional programmer aren't interested in what makes me happy in the moment or wh
Re: (Score:2)
Java and C# have endured because they were optimized for long-term development and maintenance
Doesn't hurt that they have very large companies backing them..
Re: (Score:2)
About 15 years or so ago I was in a debate with a friend as to which language was better for programming, perl (there's more than I way to do it) or python (yes, and there is a right way to do it!)? My argument came down to this: you can read python and have some chance of understanding what the code is doing.
My friend argued that the same was true of perl. I shook my head with a sad, slow motion and said "No. Here's why you're wrong. Here's a perl script; tell me what it does:"
http://99-bottles-of-beer.net [99-bottles-of-beer.net]
Re: (Score:2)
Not a good example since it's intentionally obfuscated. Perl shorhands are useful only when used sparingly not as the main course.
Re: That's the point (Score:2)
It is easier to write obfuscated code intentionally or not in perl, because is a language without seat belts. It's design and philosophy does not encourage clarity of code. On the contrary, it encourages people to try out multiple ways of solving the same problem with no way to quantify why one method is any better than another.
Try writing that code in python and see how hard it is to even think about how you would implement it in python, whereas in perl any perl guru could have done it. THAT'S THE POINT.
Flawed Methodology (Score:5, Interesting)
Visual Studio magazine argues that the biggest surprise may be that the 29-year-old language classic Visual Basic is still in the top 20 â" since its last stable release was 22 years ago, and by 2008 it was finally retired by Microsoft. "VB6 just refuses to go away, achieving cult-like status among a group of hard-core supporters."
Or, you know, the methodology being used is flawed.
TIOBE uses search engines and counts the results. It ends up counting everyone who lists it as skill to pad out their resume, every course that's offered even if nobody has ever taken it, every video tutorial that's been uploaded to YouTube and so forth. VB6 doesn't have a cult-like status; it has a lot of detritus left behind which the stupid algorithm interprets as popularity.
Re: (Score:2)
It would serve them better to consider some additional context rather than just “how often did Language X show up in search results?”
If they had done so in this case, for instance, they would’ve learned that 99.9% of the search queries were some variant of “Visual Basic 6 - what the hell were those people thinking?!”
Re:Flawed Methodology (Score:4, Interesting)
Speaking of context, think about why VB, not just VB 6, was so successful.
Remember, even in the early 90's it was possible to get a job writing or maintaining software written in BASIC. It wasn't exclusive to kids making games on their home micro. There's even a store near me that still runs on software the owner wrote, in BASIC, on his Commodore 128! This fact is important, and easy to forget. It's also a language everyone knew.
Also in the 90's, you have the rise of Windows. DOS software was on its way out, and while it was still pretty common through most of the decade, it was a clear dead-end. Things were really moving fast. Try to imagine if Windows and Linux had suddenly became old-hat in 2017 and world was abandoning them for some new thing. That's what it was like in 1995. You needed to move on and move on quickly.
What do you do? You have just a few options. Realistically, just Visual C++, Delphi, and Visual Basic. Sure, you had things like FoxPro and Toolbook, but they weren't really for serious work. They were all expensive.
Visual C++ was a punch in the face. There were tons of new things to learn just to get a window on the screen. (This is probably also the first time you ran across the concept of events!) It wasn't exactly friendly. You probably didn't have an internet connection, certainly not one at work, and definitely no stackoverflow or MSDN to help you out. Just a few not-yet-dusty books of dubious quality. The compiler was slow, and mistakes could be dramatic.
Delphi was nice, but no one used it. Well, except that one guy who still won't shut up about it.
Visual Basic, however, was absurdly easy to use. You could get up and running in an afternoon of tinkering. The interpreter was slow, but made debugging easy. Applications were fast enough once compiled anyway. The online help was actually useful. It was no substitute for VC++, but it was perfect for such a broad range of applications that it didn't matter.
VB was perfect for all sorts boring things. There were CRUD apps. You know the kind. Little more than a friendly front-end to a database. Others were the GUI equivalent to a shell script taking in a few parameters and performing some task. Some did nothing but generate reports. Still more were equivalent to config files. Fill out these fields, check these boxes, generate the config file you used to edit by hand!
These weren't sophisticated programs, but they needed to be written. It made absolutely no sense to use Visual C++. It was WAY too complicated and development took way too long in comparison. A good programmer with VB could easily out perform a good programmer in C++ when it came to those kinds of programs.
Yes, it was so easy that you could hire a kid with virtually no experience for pennies and you could expect them to make productive contributions.
That's probably where the hate comes from. Kids who didn't know any better making a mess some other guy needs to clean up. Just imagine the mess they would have made in C++! At least with VB, you had a chance.
Re: (Score:2)
I was one of those kids when I started my business 30 years ago, and it’s exactly as you described it. I ended up writing a whole mini ERP package in MS Access with VB.
You could even run complex purchasing sessions where at the end you hit a button, and the software would fax out all the purchase orders.
No way I was doing that in C++, where you needed magician status to make a text box appear. C is still my go to language for embedded, but C# is the new VB6 for line of business.
Now get off my lawn.
Re: (Score:2)
Please don't use R (Score:5, Insightful)
R is such an absolutely broken language that one of its leading guide books is called the R inferno
https://www.burns-stat.com/pag... [burns-stat.com]
R is what Perl and PHP were in the year 2000s. It has fundamental design flaws that make it impossible to create proper code that scales. I work with R every damn day and the amount of effort I have to go through in order to defend myself from the language is insane. Just to give you an idea of the insanity, is.integer(5) returns FALSE and the language has _three_ completely incompatible object oriented systems, S3, S4 and R6.
Do not use R. Port existing R libraries to Python so that we can finally bury this nightmare.
Re:Please don't use R (Score:4, Insightful)
Re: (Score:2)
That is the other big problem with R. It's basically built around a single company, the one that does RStudio. So they are free to create their own, often broken utilities and you have no choice. When you report bugs, they generally don't care, complain that they want a full "reprex" for even trivial bugs or feature requests, or claim that it is not a bug, it's a feature.
I personally got tired of asking them anything. I am mostly developing my own tools in house, in python, because nothing in the R world is
Re: (Score:2)
Re: (Score:2)
yea thanks, op lost all since of credibility just by saying the P word
Re: (Score:2)
Are you referring to the old, obsolete Python 2?
Re:Please don't use R (Score:4, Interesting)
I only got a few pages into that. Does it get better than "here's a case of how R gets math wrong, oops, it was right in that case" and "do not confuse numerical error with error"?
For those of us who are know something about computers, a disclaimer that the first chapter (or however much) is just a rant blaming R for floating-point numerical errors would be helpful so we can skip to the part that might not be a rather silly way to frame a problem inherent to modern CPUs.
Re: (Score:2)
Exploring a bit more:
Chapter 2 is about growing vectors in little bits.
Chapter 3 is about how to vectorize operations so the interpreter doesn't get involved so much.
Chapter 4 is about tradeoffs in using existing functions rather than open-coded loops. This would be helpful except it avoids specifics, opting for vague general statements like "If you have a substantial amount of data and speed is an issue, then
may be a good solution." "May be"?
Chapter 5 recommends writing functions rather tha
Re: (Score:2)
- R has no import. everything is shoved into the same global namespace with no namespacing.
- debugging it is also close to impossible. tracebacks often are wrong, incomplete or don't tell the whole story.
- the development utilities are approximate, often broken and completely irrational. there is a quite fundamental linting package that does not do anything unless you export the variable NOT_CRAN to say that you are not running on the CRAN (the library repository) build machines, and it will completely succ
Re: (Score:2)
You will get no argument from me. I try to use R every few months for something that it should be extremely well-suited for, and am always frustrated at how long it takes to find the standard functions or CRAN modules that do what I want, and further frustrated at inconsistencies in argument-passing and -naming conventions.
There are lots of good things to complain about in R, and lots of issues that show up in R just like they show up in other languages. But "The R Inferno" tries too hard to be cute, and
Re:Please don't use R (Score:5, Interesting)
R and SPSS are used a lot in scientific and academic circles. Even though you are obviously unaware of who needs this kind of environment, it is still important. That's why it shows up on the TIOBE list.
Re: (Score:2)
R is what Perl and PHP were in the year 2000s. It has fundamental design flaws that make it impossible to create proper code that scales.
You don't understand the use case for R. It's not targeted at general use, it's for people who want to analyze data.
And people who want to analyze data have no more rational choices?
Re: (Score:2)
To some extent, they don't. R benefits from the MatLab effect. You might have learned it (and only it) in school, all your statistician friends use it, and so you do too. Or maybe you need to use some obscure statistical thingamabob that has an R module written by the guy who discovered it. He wrote it in R because he's an old school statistician who learned S in the seventies. Or maybe you're some hapless grad student who heard that R is the thing to use for stats, so you do. Nobody bothered to tell you t
Re: (Score:2)
There is a lot of data analysis outside of traditional "hard science" fields, for example anthropology, nursing, social science, and library science. The people working in these areas collect data and need to find out what it implies. R/SPSS are often used in these settings.
Python assumes a level of coding skill that is not needed for R or SPSS. SciPy requires even more in depth programming knowledge. Some peo
Re: (Score:2)
Re: (Score:2)
R is what Perl and PHP were in the year 2000s. It has fundamental design flaws that make it impossible to create proper code that scales.
Perl gets a bad rep by some people who abuse it and people who just repeat hearsay and have not properly worked with it. Its "design flaw" is that it allows you to do many things that are bad, and so people can easily write unreadable code. I put "design flaw" in quotes because if someone gives you a rope you can use it for many other things other than hang yourself. It definitely has no "scaling issues" and I have worked with many Perl devs who produce "proper code" that is a delight to work with.
Re: (Score:2)
which Perl are you talking about 5, or 6?
Re: (Score:2)
> which Perl are you talking about 5, or 6?
6 is in a psychedelic retirement, renamed to something people will forget. May they one day figure our what their language is.
Perl 5 and Perl 7 are the current survivors.
But it sounds like 7 will just be 5 + Moose + a bunch of standard 'uses' and the developers are very angry about indirect object notation.
I updated an old DNS server the other day and 'fixed' it to use normal object notation. Better, actually.
if you dig +short my.ip @outsideip.net you'll talk t
Re: (Score:2)
I really didn't know... so Perl 7 announced last month and will be 5 + new things... probably as it should have been done 20 years ago. I'd agree that sounds like the right path, don't know if language has lost huge mindshare with the swiss army knife with fold out bowling ball, toilet, back scratcher and blender that was Perl 6.
Re: (Score:2)
Thanks. This explains why this is so popular with non-programmers, like Excel is as well.
Re:Please don't use R (Score:4, Interesting)
People don't use R for the language. Even those who use R know that it is one of the worst programming languages ever used, which just happens to have some of the best statistical libraries ever written.
That's the problem. For all of the talk about supposedly moving to Python or whatever, even Python's statistics libraries are not in the same league as those of R.
Re: (Score:2)
Do not use R. Port existing R libraries to Python so that we can finally bury this nightmare.
For all of the talk about supposedly moving to Python or whatever, even Python's statistics libraries are not in the same league as those of R.
GP explicitly addressed that issue.
Re: (Score:2)
GP's suggestion was effectively to try to convince statisticians to write Python code. Good luck with that.
Why not use Octave? (Score:2)
If things like R and Matlab are that popular, why are not the FOSS Octave or FreeMat Matlab clones also popular?
Re: (Score:1)
Re: (Score:2)
For any new data project I would use Python (with numpy, scipy, and matplotlib). It is very good at visualizing things,
The visualisation in matplotlib is only so-so. MATLAB is not only much more forgiving but also much more capable, especially for 3D stuff. For 2D stuff, Octave does a great job for me because it has a GNUPlot back end. It's a bit odd and often not that pretty, but it's very fast so I don't have to worry about essentially hanging the session by trying to plot a gigabyte of data.
Re: (Score:2)
R is a free clone of S-Plus. It's as if Octave became so popular that nobody under the age of sixty really remembered MatLab.
The problem with Octave is that engineering professors are conservative, universities have too much money, and engineering firms make too much money. So you learn MatLab then you keep using MatLab when you graduate because dealing with the idiosyncrasies of a clone aren't worth your time.
Re: (Score:2)
Every language has its own set of "infernos." You tend to forget about the gotchas you went thru while learning your current favorite language.
I happen to be comfortable with R, but I will happily switch to or add Python when they (python lang devs) wise up and use visible characters as delimiters. Yes, I know about Black and YAML; that's not the point. Leave white-space programming to the eponymous language.
Re: (Score:2)
The funny thing is, most of the useful bits of R are pretty simple math, and therefore pretty easy to implement in whatever language you like. Unfortunately the horrors of the C/Fortran implementation of the R module make it halfway impossible to just clone it; you have to go back to the original papers and implement the actual math.
Re: (Score:1)
Better still, just go straight up ANSI C. You might end up having to create all kinds of libraries and sub routines, but it scales without question.
Re: (Score:2)
It has fundamental design flaws that make it impossible to create proper code that scales.
That's not what R is supposed to do. R is just a fancy spreadsheet.
VB6 (Score:4, Informative)
It may offend some folk, but VB6 was a really good language for churning out classic business applications. We wrote a number of niche ERP applications in VB6, with decent success.
I don't get the hate for VB6. VB5 maybe. For C pro (Score:2)
Agreed. Mostly I hear people hating on "VB", this article speaks specifically of VB6. I've been told that VB6 was greatly improved over VB5. That may be true - I used VB6, never VB5. VB5 might have been garbage as far as I know. For me, 6 was a quick and useful way to write GUI applications.
Maybe if they only language one had ever used was C and you tried to switch to VB6 it would feel very different, and therefore strange. The language was certainly not C, and proud to be different from C.
Re: (Score:2)
It may offend some folk, but VB6 was a really good language for churning out classic business applications.
It was a really available and accessible language for that, but it wasn't necessarily a good language for it.
Re: (Score:2)
Its now on you, who is claiming that the evidence is misleading, to provide evidence.
What Microsoft got right in transitioning from BASCOM/QuickBASIC to are two things:
A native decimal datatype explicit for holding currency to the very precision that the world had settled on, and a RAD GUI designer.
It was so right, that it did not matter that from versions 1 to 5 of VB it wasnt even a compiler.
Re: (Score:2)
VB is a great language for business logic, as is evidence by its popularity for performing business logic.
It's what was promoted by the OS vendor. Unix is automated with the shell because it's what was included. DOS was automated with batch files, Amiga programs are automated with REXX because Amiga sold it, etc. Microsoft offered BASIC for automation of office apps in Windows. I've used it myself because it's what's in Crystal Reports, although there you can choose to use it with nicer operators. It's a deeply irritating language.
Re: (Score:2)
It's what was promoted by the OS vendor.
You understand that VB 1.0 is from the DOS era, right?
Microsoft offered BASIC for automation of office apps in Windows.
Cart before the horse, buddy. VB 1.0 was for DOS.
.. that was around version 5.0 of VB.
What you are talking about is VBA
Actually the compiler came with VB4 (Score:2)
Vb6 is still good (Score:3)
Re: (Score:2)
Re: (Score:2)
BASIC isnt supposed to be a "safe" language.. thats a relatively recent reinterpretation of BASIC's position.
BASIC stands for "Beginners All-purpose Symbolic Instruction Code"
Rust is most welcome (Score:2)
I've recently been taking time to learn Rust properly (streaming the experience online even), and the more I use it, the more I see the progress it brings. I can see why the learning curve could make it intimidating, but once it start clicking for you.. wow. Cannot recommend it enough tbh.
The biggest challenge has been to slow down. Usually I could pick up a language by porting data structures I've been using for years. Rust calls the bluff - some of these structures are not safe, or need to use Rust's memo
I remember when I first started off... (Score:2)
These articles were interesting for some reason that I can't seem to figure out today...
Now they kind of come and go as "there's no point to these and please just use whatever tool works best for your job."
Language designers (Score:2)
There are so many programming languages out there ... I wonder how many people understand enough Computer Science that they can design a credible programming language? Has to be over one million people.
VB6 (Score:2)
Re: (Score:2)
Partially? I believe you can develop and debug Windows apps in VB6/Windows 10. I don't think you can run VB6 web apps - or at least not ones that use WebClasses (which were always obnoxious). I use a XP VM when I have to deal with these, as even in Vista/7 there were problems (IIRC, though it's not like I'm looking for reasons to use Vista either way).
VB6 had a number of good things going for it back in the day - particularly the debugger was really smooth. But really there's no excuse for not having yo
Re: (Score:1)
But really there's no excuse for not having your code ported to .NET at this point; I've ported a few systems for work, and it's really a simple job.
VB.NET is not a replacement for VB6 for many, if not most use cases. Removing functionality, like VB6's control arrays, does not make something better.
Re: (Score:3)
I've ported tens of thousands of lines of VB6 to VB.NET. WebClasses and forms both have some awkward bits (including control arrays) that we built transition libraries for over time, and sometimes error handling is weird - but mostly it goes really smooth. Lots of stuff that's further from the UI requires no transition at all: copy/paste, wrap in a class declaration, change a little lifetime stuff, and it's mostly done. Honestly, I'd rather port VB6 code to modern .NET than try to port, say, early ASPX p
You need years of experience (Score:2)
A bit offtopic but related, apparently there is at least one job posting out there asking for four plus years of FastAPI experience. Which is interesting because it's only been around for one and a half years since the guy developed it [9cache.com].
I'd like to twist that around a bit... (Score:1)
Re: (Score:2)
A: any language that non-programmers commonly use. So probably Excel scripts. Second choice, recently-created boutique languages that attract kids fresh out of college and also tend to get new versions incompatible with existing code.
You can write bad code in any language. I once had to deal with someone's C code where for no good reason he had used "0[foo]" instead of something sane like offsetof. It took me hours to figure out how to properly unfuck that. I think I also had to deal with his C# code too,
VB Cherokee Nation... (Score:1)
There is something magical about VB-classic. It's instant gratification. It's like learning an autoharp compared to a guitar (assuming in-tune). You push a button and instant chord. No need to learn funny fingering first: FFF.
I suspect someday something like it will return for smaller apps that are unlikely to be on phones or tablets much. We just first need a real & stateful GUI markup standard to make it HTTP-enabled. (And such doesn't rule out finger UI's by the way. Finger UI's are mostly a subset o