Microsoft boffin inadvertently highlights .NET image woes by running C# on Windows 3.11 Microsoft senior software engineer Michal Strehovský has run a small .NET Core application on Windows 3.11, a version of the OS released in 1993. Strehovský posted this achievement on Twitter, explaining that it demonstrates C# compiled to native code with few dependencies. Windows 3.11 is 16-bit, but includes a library called …
He pointed the finger at the lack of compelling cross-platform GUI technology, Microsoft's enterprise focus, and the fact that Visual Studio is a paid-for product.
No, it is crap, and only former Microsoft BASIC and VisualBasic developers are daft enough to fail to realize this ... how many of you have never touched VB ? There, you have it!
Actually for much in house Windows development the .net vs Visual Studio 6 and VB6 was a massive step back unless you changed to C#
Even today lots of stuff that was efficient and easy to do can't be done simply on .net. Also all the pointless changes in "approved" APIs and GUIs.
The .net is a mess worse than DLL hell. Try installing the Microsoft Keyboard Layout editor from Microsoft. Doesn't work with ANY version of .net available from MS downloads, unless it's hidden somewhere.
Surely without backward compatibility there isn't much point to .net or Windows?
Try installing the Microsoft Keyboard Layout editor from Microsoft. Doesn't work with ANY version of .net available from MS downloads, unless it's hidden somewhere.
Apparently uses version 2.0.50727.9148 on my machine, and looking at the release date should work with any of the .NET 2.0 releases. Installed and ran without any problems.
Just tried installing it myself on Win 10. I also had no problems, and apparently am using the same version of .NET (2.0.50727.9148).
Edit, I was using this version
I don't HAVE .NET (2.0.50727.9148) and can't find it on MS site.
There are a bunch of others from 3.x maybe. Which I have. Install from that link complains and fails because I don't have 2.0.50727.9148, or any 2.x at least not on this laptop.
Found this via Google (not findable on MS Search)
https://dotnet.microsoft.com/download/dotnet-core/2.0
Slightly baffled which thing Keyboard layout editor needs. Why don't they also offer "bundled" downloads with all dependencies, like Linux Software Manager?
'simply ticking the box labelled ".NET Framework 3.5 (includes .NET 2.0 and 3.0)" in "Windows Features"'
Ha Ha!!
Good luck with that. You're far more likely to need to do this if you need .NET <= 3.5:
https://docs.microsoft.com/en-us/windows-hardware/manufacture/desktop/deploy-net-framework-35-by-using-deployment-image-servicing-and-management--dism
"lots of stuff"
"worse than DLL hell"
"Try installing [some obscure, important-to-me software]"
I think the hyperbolic, rhetorical, and anecdotal evidence taints your argument.
"there isn't much point"
The point to .NET is efficiency in the cloud without being tied to Windows, which translates into real money in your pocket.
It's true that Microsoft has shifted away from Windows being a dependency for their other big (paid-for) products, including SQL Server and Office. Despite that, it's the best desktop OS by many different measures, even if the choice of supported apps is driven by usage metrics.
This is what Trump doesn't understand about economics - if there are two parties involved in an arrangement they can both benefit. It's not a black and white win/lose scenario. You as the developer can win, as can Microsoft; they are not mutually exclusive.
You're not too far off, actually. A lot of the BIG banks (so maybe not TSB) use something almost as cryptic as InterCAL, but for serious-arse BIG work and for real.
which is based on, has not dissimilar syntax to, APL.
which can get as cryptic as greeble if you're using matrices the way it intended. Or even just basic stuff -- it uses groovy greek symbols as part of everyday life.
eg, rather than laboriously type out "split" when creating .csv files, you instead type the much more concise: "','(≠⊆⊢)".
eg, here is the entire Conway's Game of Life:
⊢world←2 2 2 2⊤0 12 5 2 4 1
0 1 0 0 0 0
0 1 1 0 1 0
0 0 0 1 0 0
0 0 1 0 0 1
{↑1 ⍵∨.∧3 4=+/,¯1 0 1∘.⊖¯1 0 1∘.⌽⊂⍵} world
1 1 0 1 0 0
0 1 1 1 0 0
0 1 0 1 1 0
0 0 1 0 0 0
Pool poor widdle Bobby wobby is threatened by the facts yet again
I know you're to dumb to see facts, but I'll give it a go.
There are also campaigns to get more MEN into nursery work, become primary school teachers and nurses
The are also campaigns to get more BLACK people to become blood and organ donors.
But you have such a huge fucking chip on your shoulder about the poor hard done by rootin tootin redneck community you completely ignore facts that are right there if you get off your paranoid Lilly white ass and look for them
The facts?
You think you have any understanding of the facts?
Fact is, you're judging people by their race, their sex, and their age, not their competence as developers, their work ethic, their character, or any other such thing, and you're patting yourself on the back for doing it. Will you ever stop looking at people as members of a race, sex, etc., and realize they're individuals, not just examples of some group you've placed them in? Bigotry is bigotry, even if it's directed at the group that it's "in" to hate at the moment. Simply changing the definition of "them" doesn't address the problem of seeing everything through the lens of "us" and "them" is ; in fact, it validates the idea that people should be judged by demographic means beyond their control rather than anything that really matters.
This would be good an well if we were all born equal and on the same footing. The 'facts' say otherwise, the reason white men are in the firing line is because we have held the cards for a millenia. This is not hate, this is just looking at facts. People that are looking at the bigger picture are seeing this trend and just because some white men can't handle losing a little bit of power and share it around doesn't define it as hate. This is just the new reality for us white guys. We need to get used to it. This is far from bigotry this the great equalizer happening right now. In risk of labouring the point, the observation in the article is that these guys were all 40ish white men (like me), if it had been 20 something Asiatic men and Asiatics had dominated the planet for the past 800 years then this would have also been a valid point, however, this has not been the case. All commentary needs to be placed in historical context. We should just take criticism of white men like men and have a thick enough skin to not see every criticism as racial 'hate'. I recognise that, of course, we are all individuals and have different life experiences and that many white guys have had a tough time in their lives. This particular obervation of white men does not diminish this, it simply observes that those that are in power are dominantly white (and men).
Part of the problem is what Dilbert called something like "automatic application of the corollary".
Eg. Most Corvettes are owned by males aged 35-55. Therefore most males aged 35-55 who own cars own Corvettes.
See the problem?
Here it's "Most of the time in most situations the power holders are mostly white men. Therefore most white men are power holders."
It's amazing how many people rake this mistake.
Yes, except that no-one says any such thing. Plenty of people misreport, misrepresent or just project their own strawman version of some notion. Most usually I think this is done in good faith; in the example above I'm sure you do believe there are "SJWs" who assert that all white men are power holders".
I'm guessing you're think of privilege, which is a different thing altogether. If you're not black, for instance, you have the privilege of a much lower chance of being denied employment, having worse health, a lower life expectancy and so on and so obviously forth.
If Historical context is important, how far back should we go? 5 years? 10 years? 200 years? How about 20,000 years? Should we consider the Mongols as receiving just deserts for their occupation of most of northern Europe? Should we despise the Egyptians for monopolising such advanced construction techniques at the price of so many slaves? Should we perhaps discuss the rise and fall of nations in the Americas, or Australia instead? Should we consider the extermination of the Denisovians, Neaderthulls, or Homo Habilis? How far back should we reach, and for how long should the descendants pay? The structure of your argument has holes that cannot simply be hand waved away.
Also White Man? You are making fun of a genetic condition caused by poor nutrition, as farming was imported, and even forced on the local populations in Europe by oppressive mediterranean colonisers. The nutrition was so poor that the local populations had to produce extra vitamin D, or die. Pre-agricultural Europeans were all Black, so how far back in time are we reaching for historical context?
"Equal" does not mean being the same, otherwise you and I would have to be genetically identical, and you would not be allowed to like corn fritters, as I don't. "Equal" means living by the same rules and standards. A King, A Priest, and a common man are not allowed to steal property, Period. The King cannot take it from the common man even though he has an army and and the common man does not. The Priest cannot seize property in the name of whatever god, regardless of if he sees you as a believer or an heretic. And neither can the common man take property from the King, or the Priest, even though the common man likely has less. Seeking a system in which every category has "equal" representation also means seeking a system in which the King, the Priest and the common man live by seperate rules. What odds will you give that the Kings rules will be better than the common mans rules?
Also this has zero to do with power, 40 year old white men have spent some 22+ years on their craft. This makes them competent, not powerful. Anyone who has spent the same amount of time honing their skills should be respected for those skills, and that respect may place them into positions that wield power (but this is by no means their right). It is their competence that grants them an opportunity to weild these privileges. When anyone is placed into such a position, without having first earned the competence, expect people to complain. And if those who have the competence are predominately from some given category, it stands to reason that most complainers will be from that given category. So when a 40 year old White Man from a community mostly comprised of 40 year old white men, should complain about someone being placed in a position of power through a fiat process that never considered that individuals competence, their admonishment is not moaning. Their admonishment is an alarm bell. Many such alarm bells are a warning that the social structures, that are meant to constrain power so that its use is limited and exercised with competence, are being usurped by those without the experience, nor the learned restraint to wield those powers safely. The incompetence increases the chance that they will maintain their position through force of power, rather than by fairly achieving the necessary competence and maintain their position due to their own merit. Personally, I would rather have a competent person in charge, not a forceful one. A competent individual would recognise my own competence and help me. A forceful individual would fear my competence and punish me.
I never did understand that move. If you want people to use your language, you have to give them the interface to learn it. The fact that MS made Visual Studio not a free package is just another gun->foot moment in my mind.
Traditionally, all programming environments have been free. Turbo Pascal didn't make you pay for its dev environment, nor did Borland ever do that. They understood that you need to incite people to come and discover, and if you set a price for entry, it just becomes a barrier that only those who have to pay will do so.
So it's no use complaining that people aren't using your language, Microsoft. You're the one who is keeping them out, for the sake of a few more bucks to throw on your bottomless pile.
@NerryTutkins
Microsoft is happy for you to spend your time and energy getting good at their environment/tools. And when you've made that investment, they want you to pay for the privilege of using their tools to produce software which helps Microsoft sell their OS. I can't speak for anyone else but that tastes a little bitter to me.
What exactly are you expecting? "For-profit company tries to find ways of making profit" isn't much of a surprise surely?
It's the same approach for the Unity and Unreal game engines (which between power a lot of modern games), free for personal use, and the cost goes up as you make more money.
There's always the old Adobe approach I suppose, where you turn a blind eye to low level piracy of Photoshop/Illustrator etc. so that young designers grow up (pirating and) using your products, so that when they get a proper job, their employer ends up having to shell out for a license. Of course these days they go for the "charge everyone all the time" tack.
The young devs are all using VS Code, whether with .NET or not, and slowly being indoctrinated into the Way of Things. Either that or they will learn to hate hitching their ride to FOSS and having to figure stuff out all over again every time they switch jobs, or every two years, whichever comes first. While giants like Apple, Oracle, Google, etc. do cancel projects, it’s much less likely than with independent devs.
"Traditionally, all programming environments have been free. Turbo Pascal didn't make you pay for its dev environment, nor did Borland ever do that."
Ermmm, yes they did. You might be thinking of that version you copied off your friend.
Here is a Q and A I just found from 1998, including the prices https://community.embarcadero.com/article/technical-articles/162-programming/14734-borland-c-40-general-product-information.
If you used any of the graphics libraries for dos it wouldn't have been. That was also an addon at cost. The C / C++ course i did, didn't notice that the version they had didn't have it, so removed the course work that included it, after I had completed it at home using my fathers version that had them.
I also remember Borland J builder, used that through university, UML creation, got that off PC Format, might have been computer shopper CD version i believe.
But it still doesn't change the fact that it was not free. You could get older, limited version, just like you can with visual studio (just limited) for free.
Free development tools under Windows appeared especially with Java, and especially after Sun open sourced NetBeans and IBM after failing badly with VisualAge open sourced it as Eclipse to lure people on making it usable for IBM expensive enterprise products.
Before you had commercial solutions from Microsoft, Borland, Watcom, Sybase and probably others.
Sun gave away for free the Java vm, libraries and compilers so it made possible to code only the IDE.
And of course under Linux you got all the development tools for free.
That was well into the nineties, a quick Wikipedia says it was actually 2000 when it was open sourced. At that point Windows had been out for fifteen years.
Sure, gcc was free for Linux, the issue was that compared to Watcom (which was cross platform and a reasonable price) it was pretty shit.
I was mad enough to develop a custom FTP program for Windows NT using Watcom under OS/2. It worked fine and wasn't that difficult. It would have been considerably harder fiddling around with gcc debuggers. Even now they're not particularly amazing, especially with respect to live watching of variables (I've tried a few, they're all a pain). Under Windows such features have been standard for decades, and they're available freely in windbg (and some of the Visual Studio offerings if you're not developing for an organisation with more than five people in it)
especially with respect to live watching of variables (I've tried a few, they're all a pain).
Really?, I've never found watching live variables in kdevelop an issue. Has always worked fine for all the projects I've worked on. You are compiling with the appropriate switches?
Borland knew exactly how to charge a very high fee for its IDEs and compilers. Turbo Pascal was most definitely not free and I recall Delphi costing over £1k for a fairly basic version.
Also the dev mentioned in the article and its author are not well versed in .Net or Visual Studio since there has been a free Community version for several versions now. You can even commercialise the software you make with it up to a certain value.
I remember in the late 1980s paying sub-£100 amounts for my own copies of Turbo C and Turbo Assembler (I used Turbo Pascal too at the time, but work paid for that so I don't remember the price), but I have some recollection of there simultaneously being a 'professional' version that was somewhere closer to £500.
Borland/Inprise/Borland/CodeGear/Embarcadro prices increased steadily from TurboPascal days, which first versions costed $49 or $99, I don't remember exactly. Today they ask a yearly subscription which is at least a four digit sum - as they are trying to offset a dwindling number of developers with higher prices.
Ironically, most compiled language were relegated to niche uses but maybe C/C++, and now we got manage languages compilers to achieve higher performance....
"Turbo Pascal didn't make you pay for its dev environment, nor did Borland ever do that"
Apart from confusing the product with the company, my copy of Byte from 1989 shows Turbo Pascal going for £99, or £169 for the Pro edition. They did do a free version later in the Windows era, but that had a limited licence and Microsoft had a similar edition.
Microsoft's prices are certainly a bit insane.
I think that QuickBASIC 4.5 cost me ~£130 back in the early 90's. As a hobbyist, it was perfect. I should stress that this is the full QuickBASIC that could produce compiled .exe files, not the cut-down interpreted IDE that shipped with MS-DOS 5 and higher.
The alternatives were PowerBASIC (which I eventually moved to) or... Turbo Pascal, I guess?
Actually, the real alternatives were things like the C compilers, which I think cost more like £300 and upwards. You could start cheap with Borland or Microsoft C, but if you wanted something like Watcom you were going to have a very light wallet afterwards.
So what that purchase of QuickBASIC 4.5 got me was the ability to write little programs that I could share with friends without requiring any runtime or installation. (This was DOS, though, so that's not saying much.)
A decent dev environment costs money. I understand that. Compilers or runtimes take time and effort to develop. But free software has been going for over 30 years, and has accumulated millions of man hours of work - it's at least 90% there.
I just checked the prices of a Visual Studio Professional subscription. $1,199 for the first year, $799 renewal after that. I understand that you get a lot there - Azure credit, dev/test licenses for some Microsoft software - but that's still a very steep price.
By contrast, a couple of years ago I treated myself on my birthday to a subscription for all the Jetbrains products. All of them - IntelliJ IDEA, GoLand, Rider, PyCharm, Datagrip, and more - and it cost me £200 a year. It went down to £159 for the second year, and will be £119 a year this year and onwards. That's a lot of IDEs, and a lot of expertise.
For £119 a year I can write in Java, Go, Python, Ruby, C#, C/C++, Ruby, SQL, JavaScript or PHP. That's great, because if I do get a head injury at least I'll have options. (Insert your own joke about which ones require a head injury.)
If Microsoft wants to attract new developers, they need something that's down at the £250 mark.
Visual Studio Code doesn't cut it, by the way. It's great for PowerShell, and not too bad for C# - right up until you want to compile to an executable. Getting Visual Studio Code to do decent compilation is a bit of a pain.
Certainly more of a pain than anything from JetBrains.
If Microsoft want more developers using C#, they need to drop their enterprise-style pricing and make Visual Studio much more attractive. I know that there's a Community Edition, but the cost of the jump from free to non-free is incredibly high, it's no wonder everyone just goes off and uses something else...
"If Microsoft want more developers using C#, they need to drop their enterprise-style pricing and make Visual Studio much more attractive. I know that there's a Community Edition, but the cost of the jump from free to non-free is incredibly high, it's no wonder everyone just goes off and uses something else..."
It's the same as travelling in business class. Way too expensive, but you don't care as you're not the one paying it.
There is a big difference - the old tools (and VS today) included in-house developed compiler/interpreters, libraries/tools and the IDE. That was (and is) far more expensive to develop. Even when some libraries/tools were licensed, it had a cost.
Today you can get an IDE which is built to use compilers/interpreters, libraries/tools built elsewhere and usually open source so no licensing costs. That is a huge saving in development costs.
There are downsides - often you get a single compiler/interpreter for a given language, and less competition means also less innovation.
Moreover languages are now multiplicating like rats, and a very fragmented development world will soon become a maintenance nightmare.
"Turbo Pascal didn't make you pay for its dev environment"
Yes they did. At school I remember there were unofficial (ie. the school did not know nor approve) 'liberated' floppy disk copies being exchanged between us IT students for use in home assignments. We wouldn't have done that if it were freeware.
don't forget THIS gem: "C# remains popular, ranked sixth after JavaScript, Java, Python, PHP and C++"... forgetting the fact that ".Not" and/or "C-pound" are less than 10% of the total amount of development going on, and THE MOST POPULAR lingo (Java) just happens to be cross-platform...
though the current group of free IDEs really disappoint me (IntelliJ, Eclipse, etc.) there are some people who like SlickEdit [I've seen but not used it, long ago] and I just use Pluma and the command line tools.
Recently a friend of mine, who was a ".Not" developer some 10+ years ago, sent e-mails around about finding work. My suggestion: learn cloud and Kubernetes, something with an actual FUTURE.
"forgetting the fact that ".Not" and/or "C-pound" "
Why DO you KEEP calling IT c-POUND, when # IS called SHARP. you DON'T say THAT note THAT was PLAYED was A c POUND, no YOU say IT was A c SHARP. the POUND sign IS £. OR if YOU mean POUND as IN weight THEN lb.
God that going to uppercase annoyed me while writing it.
Dude, have a little tolerance - older people often have difficulty adjusting to newer things, like programming languages, having to use different input devices (mice), changes in visual styles (2D vs 3D), or tags that allow you to emphasise text without having to resort to capitals or adding stars/underscores/exclamation marks around words, and so on.
Changes can be scary for people who are used to things being a certain way, and this fear can evoke other strong feelings.
Just smile and nod, and try to empathise a little - after all, we wouldn't want some social justice warrior getting on your case for being ageist now, would we?
Older people often have difficulty adjusting to newer things, like programming languages, having to use different input devices (mice), changes in visual styles (2D vs 3D), or tags that allow you to emphasise text without having to resort to capitals or adding stars/underscores/exclamation marks around words, and so on..
Just because it's 'new' doesn't mean it's better. When you get older you'll come to realise that in more ways than you can possibly imagine right now.
PS emphasized 'new' without markup because it's easier and quicker.
"When you get older you'll come to realise that in more ways than you can possibly imagine right now"
When I get much older, I'll probably be too dead to care, to be quite honest...
I'm in my mid 40s, and computers have been part of my life since before I turned 10 - as hobby, then degree and now over twenty years commercial experience. I've seen a lot of change - not as much as others on here, I'll grant you, but I am no stranger to it. The key difference here is that I am not afraid of change, regardless of whether I feel it is better, worse or has no real impact.
I accept fully that others may not share this lack of fear - heck, I was the black sheep for my entire time at secondary school as I embraced this "computer fad", which was threatening to disrupt an old and very traditional - and therefore fearful of change - establishment.
But change is inevitable, for without it, we wither and die. And, as I said, it's not always "better" - but that's just my opinion, which I will voice in what I hope is reasonable and rational debate. So, for example, I would agree that the single quotes style of emphasis is easier and quicker, but not as immediately obvious as using a tag - however, it's not overly intrusive on the eye when reading it so... no real impact, no fuss.
However, excessive and continual use of practices that many - not just myself - see as the typed equivalent to a five year old continuously yelling "I am right, listen to me!" will incur a response that will gradually descend into ridicule and even post-reporting when the content of the post start impinging on the forum comment guidelines.
Growing old is mandatory, growing up... not so much. And unfortunately, there are those who, regardless of their feelings about change, seem unable to express themselves in an adult manner.
Also, the pound symbol looks like £, not #
# is Hash
♯ is Sharp
It is OK to use # as a substitute for sharp as it looks basically the same and is easier to type
We had APL (A Programming Language), A, and A+
We had BCPL (Basic Combined Programming language), and B
We had C, and then C++
We had the next note in the musical scale C♯, though no B♭ or A♯
followed about a year later by D
E came before C♯ and D for whatever reason
F came a few years ago, but not before F♯, so everything is now completely out of sequence
G came before any of the other ones, but that's OK if you are using a G major scale.
As a former music student, I ALWAYS saw C# as C-Sharp, BUT SOME PEOPLE might now see it in other ways way, especially those Brits who see that as a version of a weight-describing Pound symbol!
Now in terms of UPPERCASE and LOWERCASE, I do understand Bombastic Bob is almost AS ANNOYING as ME but that is neither here nor there! We're both aged white a$$ed programmers/IT people who most-definitely enjoy OUR DIVINE RIGHT privileges and can most-rightly tell off those STOOOOOPID P.C. (politically correct) pri&*s to go SOD OFF !!!
We'll fight tooth and nail for our TAKEN RIGHTS --- A "Right" isn't EARNED --- IT IS ALWAYS INITIALLY TAKEN BY FORCE and then passed down through the generations by lordly, gentrified MEN of odiferous honour who can count upon our LOOOOOONG-held familial wealth, well-oiled Oxbridge educations and fiercely PASSED-DOWN political connections that let us STAY ON TOP of the wealth and political pie as we so RICHLY DESERVE, while AL YOU pathetic peons can lament and WHINE to your heart's delight as you pathetically muse upon subjects and life choices you had no hand in!
Sooooooo, the middle finger to you all and NOW GO GET ME MY COLD BEER !!!!
.
> those Brits who see that as a version of a weight-describing Pound symbol!
You are confused. The 'Brits' do not use 'hash' as pound weight. It happens that on a US keyboard the 'hash' is shift-3 while on a UK keyboard that is where the pound money symbol (stylized L) is. The use of hash as a form of 'lb' (pound weight), according to Wikipedia, is chiefly in the US.
https://en.wikipedia.org/wiki/Pound_(mass)
Traditionally, all programming environments have been free. Turbo Pascal didn't make you pay for its dev environment, nor did Borland ever do that.
I distinctly remember shelling out some big ones for the privilege of my own copy of Turbo Pascal for a programming course I was doing in the mid-nineties. The "free" versions of Borland's Turbo $DEVENV were either crippled, or pirated - mostly pirated.
Microsoft was driven to release its Express and later Community versions of Visual Studio to try to stop the haemorrhaging of Windows developers to Linux and FOSS community. I'm still expecting them to release the source trees of decently long-dead Microsoft OSes so us hobbyists can fool around with them the way we can with Linux, *BSD and OpenSolaris.
Traditionally, all programming environments have been free. Turbo Pascal didn't make you pay for its dev environment, nor did Borland ever do that.
Nope. I'm not sure where you got this idea from, but the advent of free development environments is a pretty recent one. The examples you give are just not correct: Turbo Pascal was a paid product. The wikipedia Turbo Pascal page says "Borland released several versions of Turbo Pascal as freeware after they became "antique software" (abandonware), with 1.0 for DOS on 1 February 2000, 3.02 on 10 February 2000, 5.5 on 21 February 2002, Turbo Pascal 7.01 French version in year 2000". Borland Delphi and C++ were always paid products, and not particularly cheap as I recall. This goes right back to C compilers and assemblers for earlier machines like Amigas and C64.
Compilers and languages tended to be bundled with UNIXes back in the day, but those unixes were paid products anyway. And MS gave you QBasic bundled with your (paid) copy of DOS. There was Basic bundled into the ROM of lots of older machines like the C64 and the IIe, but that was in lieu of an operating system - as soon as you got a real OS, you stopped getting free languages. Even AmigaBasic was a paid product.
The advent of compilers and dev environments being mostly free is a fairly recent one. It only really became a thing once some decent open source IDEs started to spring up. Eclipse was probably the first big major one. I've been using free tools for about 15 years and I'm a relatively early adopter. Yes, there were people using free tools much earlier than that, but it certainly wasn't widespread. I'd say it's only become the default in the last decade or so, but that is just a gut feel and I might be off by a bit.
WHAT are you talking about!?
I've used Turbo Pascal, Turbo Pascal for Windows, Turbo C, Turbo C++, Borland C++, Delphi, etc. ALL PAID FOR!
The only "free" Borland "IDE" I used was a very basic Turbo Vision based editor that came with a C/C++ book.
Nothing wrong with companies charging for software. Most are not charities after all!
Exactly! I'm pretty tolerant of neologisms and generally easy to get along with (despite what everyone who knows mw says!), but one MUST draw the line somewhere and, for me, "productize" not only CROSSES that line, but leaves it sitting in the dust a mile back, shaking its head and saying "What the hell just happened...?"!
The community version is free for individuals and small companies (up to 5 users), as well as for academic use, open source development.
I happen to be a white male in his 40s, but I am interested now in looking at alternatives where I might be able to mix with hip ethnic minorities in their 20s.
I'm almost 60 years old, I *HATE* C-pound and ".Not", but used to like Visual Studio before the 2000's when they went all "VB propery sheet" and became HARD! TO! USE! for typists - all that moving my right hand off of home row to mousie-clickie where a HOT KEY used to do the job!!!
MS's IDE has sucked too much following Visual Studio '98 where it actually made sense - well, for C/C++ devs at any rate. And that's what I do. And that's what I want. And yet I'm still forking over the $800/year for MSDN, because I need it on occasion for customers. It's a net gain, but not much gain.
The latest crop of "feelies" making BAD decisions ("feeling", not THINKING) "on behalf of us developers" first resulted in ".Not", then C-pound, and *ULTIMATELY* "The Metro" and UWP and the 2D FLATASS look.
And they want to prove WHAT now by putting their ".Not" lipstick on Windows 3.11 ???
Hey Microsoft, make sure you apply it to the end that goes OINK!
Hey Bob,
You probably last accepted advice when JFK and Harold Macmillan were in office, but I'll offer some anyway: when criticizing a product, your opinions are more relevant if you happen to have used it in the last twenty years. As an example, your first complaint regarding hot keys is simply untrue. There's a hot key for nearly everything, and if not it takes 15 seconds to set one up. But reality doesn't come into it when you are making stuff up does it?
Learning new things gets difficult as we age, and that frequently results in people using hostility to mask their ignorance and anxiety. This hostility then often leads to tedious and uninformed rants, which are almost invariably of the form: GUIs are for idiots, things were better in my day blah blah.
This brings me on to your comments more generally. They are repetitive and full of false assumptions or assertions, you are always critical of Microsoft and it's all so very boring. You're IT's version of someone who claims to be ex-SAS soldier, but never got further than the boy scouts. Change the record mate, we have heard to it too many times.
Take a look at Lazarus, it's a Free (as in speech) alternative to Delphi with all the nasty bits taken out. Software natively compiles on at least Windows, Linux and MacOSX, and since it uses the native GUI toolkits it'll always look and feel native. For all of those platforms you get a fairly large (10 Megabytes) static binary you can just drop onto the system and run.
One size fits all is a myth.
You need at least different GUI / Presentation layer for:
1) Phones
2) Tablets (even if in theory same OS as phone, or it's like a magnified phone app). Needs to work touch only, but also with keyboard / touch pad.
3) laptop/desktop OSes such as Windows, MacOS, usually touch is irrelevant even if screen has it.
4) Embedded with maybe Web Admin
5) TV / Set box / Disk player etc GUI for TV more than 1.5m away with no mouse/keyboard/touch
Also drive letters vs mounts causes an issue even with supposedly platform independent Java. You thus need different storage programming.
Some OS or form factors don't sensibly be suitable for more than one window, even if possible.
Yes and no.
While the user experience on all those devices (and others, you missed AR/VR) will be different - in some cases drastically - you still want the underlying engine to be the same so the application layer can share as much code as possible. You certainly don't want to write your application core in one language, and then have device specific layers on top in whatever language is flavour of the month on any given platform.
While you can get away with that - and that's exactly what the video player I was working on up until recently did (on Windows, macOS, iOS, tvOS, Android, Unity, React, and WebAsm) - keeping the different layers up-to-date will be a massive pain in the arse and a drain on resources - not to menion morale. (We had always planned on moving as much of the UI as possible into the C++ application core eventually.)
> you still want the underlying engine to be the same so the application layer can share as much code as possible.
True, although I'd say "as much as practical," not "as much as possible," I also really think that the user experience should not be sacrificed in order to achieve this. All of the methods I've seen in this direction so far (with WIn8+ being the poster child for this) have sacrificed the user experience.
If saving on production costs makes things worse for the user, then those savings should be skipped.
This post has been deleted by its author
> Which raises the question of what actually is the native look and feel of Windows these days?
Well actually that still is the same as in the Windows 9x era. You can see that when all of the modern GUI extensions crash. I think it even reverts to the "System" bitmap front.
Of course if you don't like the look of the GUI elements you can use the OwnerDraw event and draw them yourself.
The last time I had to use Qt (between three and six years ago), my experience was it worked differently on each platform, and failed to work in different ways on each platform. At least that was the state of the networking side of it: Write something on Linux and it didn't work on Windows. Switch to some other way of doing something and that worked on Windows, but not on Linux. The UI was consistently mediocre across all platforms we were supporting from what I remember.
Typical Microsoft bloat.
Visual Studio is a clear cumulative effort from Microsoft being too lazy to remove redundant code from their own platform, having to potentially reserve 60GB of space just for the IDE is taking the piss.
It's a clunky dinosaur IDE to support clunky proprietary languages that have to be deployed in a restrictive ecosystem from both hardware and software standpoints.
Hated working with .NET half a decade ago, as its methodology at the time was out of date, and development time was eternally hindered by complications due to inconsistencies with compulsory library inclusions.
God help anyone still clinging to that sinking ship.
I am now a little beyond the "stereotyped" c# programmer. But am required to use VS and C# as part of my day job. In my 30+ years in the fintech industry I have used everything from Assembler on the mainframe, COBOL, Clipper, VB, C#, PHP etc. etc.
I have used the Community editions of VS from around 2007 when the first "free" IDE made it's appearance and never really missed any of the "features" in the paid for versions...
.Net Core is frankly a breath of fresh air and in the bigger scheme of things - and particularly if MS can avoid the usual bloat that occurs in time for any language - it will be my language/framework of choice for sometime to come. Being in the fintech sector, the development requirements are not particularly technically "adventurous" and the sheer number of MS dev's out there means that you have a fairly deep pool to draw on from a continuity perspective.
I feel that the generalised demographic of c# dev's is just the fact that many of the younger dev's get drawn into the "cool" social media / AI projects...unfortunately the takeup of new dev's is dwindling somewhat for the last 15 years or so as well which further shapes the rather distorted picture we have of the dev industry as a whole.
The 80386 was a 32 bit device. It was backwards compatible with earlier 16 bit and 8 bit processors. (8080/8085, 8086, 80286). What took a long time to get established was 32 bit code generators and linkers for the compilers.
What this is really telling you is that you don't need many megabytes of code space and a whole tier of libraries to write a 'Hello World' type program. We've just got used to bloat because our high performance hardware and almost free storage. We should really take a step back and ask us why we need what used to be called a supercomputer to display web pages or run routine applications -- is this bloat really necessary or desirable? (Because it doesn't add to code quality, it detracts from it.)
Mostly, you're right to argue for less bloat in software's resource usage. It can be really weird to muse on what someone from a few decades ago would think of the cheapest computer we could find today. However, not all of the increase in usage is a bad thing. Some stems from techniques that actually improve the experience, either for users or for future developers.
An example might help prove the point: I used to use an audio editor which was written in the late 1990s, and ran for a very long time despite not getting updated. It had several useful features I liked, a nice interface, it was just perfect. Well, not entirely. Since this was intended to run on computers of the Windows 98 era, the developers didn't plan on having very much memory available to them. They dealt with this nicely, using only about six megabytes, no matter how much audio you wanted to work with. All the audio would be streamed from disk, with temporary files holding processed data before it was saved. This meant that, although I had gigabytes of memory, if I wanted to do any processing that required modifying many parts of the data, it would read in a chunk from disk, process that chunk, write it back out to disk, and continue. It would have been faster had it been possible to inform it that it could cache a lot more. I ran a benchmark between this program and a more traditional editor with memory caching, and found that then (around 2009 I think) having audio data in memory made compression operations take about eight times less time than it took for the program reading from disk.
In that specific situation, ramdisks were not a feasible option. The program had a hardcoded directory for storing the temporary files containing edited data, which was the user-specific temp directory. That directory was already created on disk, and therefore could not be mapped as a ramdisk. There were some theoretical options, such as writing a driver that intercepted file requests from the temp directory and wrote them to a ramdisk or trying to modify the binary to write to a different temporary directory, but those were not particularly easy compared to switching to a more modern and full-featured audio editor, especially considering that I don't work in audio editing and it was just a hobby.
The main point, however, is that there can be benefits created by using extra resources. Perhaps a better example is using any interpreted language. Merely by spinning up that interpreter, the program is almost guaranteed to run more slowly and using more memory than the same program implemented in something compiled. However, it is probably easier for the team to debug or improve upon, which creates benefits for the user. I am usually fine with it if a programmer decides that, for their sake, they will be using an extra megabyte of memory without adding a feature. It gets very annoying when they decide that a gigabyte is also forgivable.
I was hired at a place a year ago. After a reorg, I was thrown out of OSS world and into the pit of m$.
Prior to this, I had heard that C# and Active Directory were the two things that m$ got right. My opinion of the person that said this has been adjusted.
1) Documentation links were broken. Those that were not were often out of date (that is--WRONG). No appealing to source to figure out what was actually going on. I ending up black-boxing built in classes.
2) Tooling which, besides not matching the documentation, produced results which were unusable in a way which could not be corrected by scripting. Basically, the results were hints as to what might actual be usable following more black-box exploration.
3) Built in classes that throw exceptions when an array is empty.
4) Built in threading support where the m$-recommended usages is not thread safe.
As for Visual Studio...
1) Who in the !@#$ was the #$%^&*! that signed of on a %^&*!@# IDE that has to run as root?
2) How in the &*^% did anyone approve an IDE that updating required a *&^%$#@ reboot?
That's not even touching the daily usability issues, such as the contradictory terms used for the same thing in various places, the apparently arbitrary divisions and grouping of features, or the sheer impossibility of doing certain common tasks inside the tool.
Of course, if SO is a bad joke in the OSS world, the m$ stuff is truly macabre. The eighth- or ninth-ranked answer was routinely the one that technically would work. Not that I would actually pollute my code with the garbage, mind you. But it would function.
What this shows is how impressive C and C++ is as a mechanism for backwards compatibility and digital preservation.
The .NET VM is massive and yet all of that C++ code can still be wrangled to work on Windows 3.1x.
Imagine if the VM had been written in Java or Swift. It simply would not have been feasible to back-port.
It just shows, when it comes to portability; C and C++ really are the only solutions.
That would be sweet. I am not a big fan of languages that require a VM to run on and this includes Java and C#. Getting C# to run as a single (and, hopefully, not very fat!) executable would be a game-changer. Well - at least I would be no longer so hesitant to use this language ;)
Maybe I don't understand the point of this article......
*
But here are a couple of Windows 3.1 applications that run perfectly on my Intel x86_64 laptop:
- Microsoft Multimedia Viewer
- Borland Turbo C++
*
No fiddling, just double click on the appropriate EXE file....and both run like a champ.
What is the secret sauce you might ask.......Fedora 31 and WINE 4.x.
"But here's an uncomfortable truth – it's a number of orders of magnitude worse in our community. In its current form .NET is for 40-year-old white men. That's the visual.""
What the heck is with the blatant racism?
Is there any other "visual" that it would be acceptable to refer to in this way, to indicate that something is just automatically repulsive?
I can remember installing Windows 3.11 on many machines at work using six floppies and around half an hour of time.
Typing c:\win used to bring it all to life.
I can say that up until 2018 my wife used the DOS version of Microsoft Money running on Windows 95 in her business until the Accountant requested that she move to a Sage product.
From what I can remember it ran within an early text editing program going by the name of QBasic.
Happy days.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
