-
Jan 20th, 2018, 10:57 PM
#81
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Schmidt
Absolutely right.
I have to, because the RC5 is only one part (one cog in a larger "gear box").
To achieve the goal of a new VB, every part has to fit - and I have absolutely no doubt what
would happen, should I open the RC5 to the public now.
Baka for example would just love to "rip out" a few rendering classes, adding a few things -
then compiling his own "better version" (more oriented to gaming-scenarios) - and encourage
younger developers to use that instead ("look, it's much smaller than the original" - sorry Baka,
in case that hypothetical scenario is "too far off", but I don't think so).
The result:
Now we have a split code-base (with incompatible interfaces, bug-reports for one part not that easily
recognized and reflected in the other part).
Such effects will cascade - and the next young "game-developer" who is talented enough, will then
come up with "his own idea of nice gaming-widgets" (probably not even knowing, where 99% of the source
for his game-engine originated from, and that such Widgets already exist, in much better incarnations).
Then look at Eduardos comments, who's absolutely convinced that the RC5-Class-Interfaces are of "totally wrong design".
He would probably start to put RC5-Class-Code back into *.bas Modules, to be "more like original basic".
There we have the next scattering-effect - and the next set of incompatible "basically RC5-based" code-examples,
which will float around in the forums after some time.
Now, if the question arises "which GUI should we use, to implement a new IDE", some will probably
"vote" for "the Baka-engine" - some old-style guys who "never had some real use for Classes" -
will perhaps favour Eduardos approach to things, and so on.
As a result, probably nothing serious will happen - too much diversification, too many opinions
about "the right way to go about it" - well, that's the "FreeBasic having no decent GUI"-effect I was already mentioning.
In the end it all boils down to "stable interfaces" and "stable behaviour behind these interfaces".
That's what we rely on in our little msvbvm60.dll, and which enabled the community to "share compatible sources".
The new compiler will have (in addition to the original VB-Runtime-Functions) a larger, modern Class-framework
in addition (RC5-based, because that is needed not only to support "modern-stuff", but to become platform-independent) -
a new IDE which aims to be portable between Windows and Linux cannot use Win32-APIs
(besides a handful of exceptions, separated by compiler-conditionals).
So, the platform- and system-specific APIs have to be "encapsulated away" in that Class-Runtime - and
it needs to be as stable as msvbvm60.dll is today (interface-wise and implementation-wise, the RC5 aims for that).
As long as that is not understood (and it obviously isn't by the majority of the VB-community), I have to keep
the sources closed (to maintain that interface- and implementation-stability which is needed in the long run).
Do you understand what I just wrote, or do I have to "murder even more words"?
Why do you even make such deprecatory statements in a thread which discusses the survival of the language you love and currently use?
What is your plan to ensure the continued existence of that language?
Are you willing to help people, who have already invested years to ensure its survival?
If you are not willing to help, fine by me - but then have at least the common decency to:
- not make any demands
- not join with those community-members who try to actively hinder the project in a "dog in the manger" like fashion
https://en.wikipedia.org/wiki/The_Dog_in_the_Manger
Olaf
Well I was trying to be constructive, complimenting your work and discussing what I thought would make it better. You want to disagree, fine, but you're being rude. So I'll return the favor. Yes, I understand your post, you seem to believe that anyone disagrees with you is just wrong and malicious. For the 50th time, I'm not saying you'd ever have to accept 3rd party code into your official development channel, and I don't believe anyone in this community would attempt to force you to. That's what your entire post is about, and for the 50th time, it's a bad reason. "Well should we use baka's GUI???" "No, we're not using 3rd party code." End. You've not proffered any outcome that would disrupt your development unless you let it. A few random offshoots are entirely harmless.
Believe it or not, there's actually such a thing as a good faith disagreement. After looking at the justifications, as my post indicated, I felt open source would advance your project, not hinder it. Disagree, ignore me, fine, but you know damn well I'm not coming at this from an angle where I'm trying to block progress. While I can't compare to something like RC5, I have spent thousands of hours trying to extend the shelf life by showing people how to set up modern UI/shell features. Sorry if my contributions don't measure up but if you ever get down from your ego trip you'd see I am someone who is working towards keeping this language going, admires your work, and was making a good faith argument that it would be better open source. And you bring that garbage?
-
Jan 21st, 2018, 12:32 AM
#82
Re: Vb6 , the Future, and what I have discovered
Originally Posted by fafalone
I am someone who is working towards keeping this language going,
Ok, let's see if you're at least so kind, to just answer this single question:
Is it a platform-independent VB6-successor which makes more sense,
or is it a VB6-successor which remains MS-platform dependent?
Olaf
-
Jan 21st, 2018, 01:22 AM
#83
Re: Vb6 , the Future, and what I have discovered
Geez. Personally, I'd be totally happy if we had an IDE that would load, run, debug, and compile my project that was still totally tied to Windows, but was current with respect to many things like Unicode throughout, styles, interface, and many other things (just to name a couple). It'd be super cool if it was open source too, but not absolutely mandatory if I could totally trust that it would be supported and updated into the foreseeable future.
Everyone Have a Pleasant Night,
Elroy
EDIT1: And just as an FYI, that's totally where we were with the VB (and PDS before that) line, before Microsoft screwed us over.
Any software I post in these forums written by me is provided "AS IS" without warranty of any kind, expressed or implied, and permission is hereby granted, free of charge and without restriction, to any person obtaining a copy. To all, peace and happiness.
-
Jan 21st, 2018, 03:42 AM
#84
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Schmidt
Is it a platform-independent VB6-successor which makes more sense,
or is it a VB6-successor which remains MS-platform dependent?
IMO, in the short run platform dependent is ok as we are already used to it. At the long run it makes sense to compile it's sources platform independent.
-
Jan 21st, 2018, 09:42 AM
#85
Re: Vb6 , the Future, and what I have discovered
The day when VB6 programs no longer run on Windows is the day you have to throw away everything and start over anyway.
What that day looks like we don't know. Maybe it eliminates Windows and replaces it with some single "terminal" like a browser, with all code running either as client-side script or Azure-hosted Linux that only runs something like .Net Core.
The good news is that long before that day comes we'll have notice of it. We merely have to pay attention to what Microsoft is pushing at their conferences. Most people lag years behind the trends anyway and ignore these direction announcements. Look at home many kept hugging Windows 9x for so long, then jumped to WinXP and started hugging that for years. These same laggards are now squeezing Win7 to death in their desperation to stay on the bleeding edge of the past.
-
Jan 21st, 2018, 01:39 PM
#86
Re: Vb6 , the Future, and what I have discovered
I agree. If a unified desktop is about to go away, we'll have some warning. There are things that just don't work on any other platform. I currently work on a plug-in system. Third party plugins don't work well for anything served up from either the cloud or a web browser, because (currently) you always need somebodies permission to add a plugin. This could change. We could all have "personal clouds" where all of our applications reside, so we could add plugins to our own personal clouds without impacting anybody else. We may even be moving that way, but not with any world-changing speed.
My usual boring signature: Nothing
-
Jan 21st, 2018, 01:44 PM
#87
Re: Vb6 , the Future, and what I have discovered
I have been using a cheap NAS/media server and a small Windows Home Server box on my home network for years. That isn't a difficult or expensive thing, and the Windows box could even take over the functions of the other box.
The result is nearly all of the "personal cloud" you need. You can even expose it to the Internet if needed, though security becomes a perpetual concern at that point as you identify and secure against each new threat.
-
Jan 21st, 2018, 02:14 PM
#88
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Krool
At the long run it makes sense to compile it's sources platform independent.
Y'all are starting to sound like you're advocating a p-code-like solution. In terms of a compiled platform-independent language, Java is the only one I'm aware of. However, p-code isn't that far off. I know it's not terribly secure, but that could probably be improved. And, I'd bet that it wouldn't take that much work (although not trivial) to make p-code run on other hardware/OS platforms.
I've been shot down several times arguing for a p-code engine alternative to the VB6 IDE, and here it's starting to sound like a good idea.
Y'all take care,
Elroy
Any software I post in these forums written by me is provided "AS IS" without warranty of any kind, expressed or implied, and permission is hereby granted, free of charge and without restriction, to any person obtaining a copy. To all, peace and happiness.
-
Jan 21st, 2018, 09:09 PM
#89
Re: Vb6 , the Future, and what I have discovered
Sometimes, we think others are conservative, in fact, others are innovating, or to make a acme of the product (a perfect product).
-
Jan 21st, 2018, 09:35 PM
#90
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Schmidt
Ok, let's see if you're at least so kind, to just answer this single question:
Is it a platform-independent VB6-successor which makes more sense,
or is it a VB6-successor which remains MS-platform dependent?
Olaf
Platform independence requires closed source?
-
Jan 21st, 2018, 11:05 PM
#91
Re: Vb6 , the Future, and what I have discovered
Originally Posted by fafalone
Platform independence requires closed source?
As expected, you are trying to avoid answering what I've asked you further above.
Why is it so hard for some people to simply state "like a man" (as Krool did), what they think.
So please - you go first - and then I'll answer (in any detail you wish) the one you've asked above.
Olaf
-
Jan 21st, 2018, 11:37 PM
#92
Re: Vb6 , the Future, and what I have discovered
I worry about the "getting hit by a bus" scenario and the lack of future bug fixes that may follow, but not that much. Even with the source code you might not have the necessary tools or expertise to recompile the thing, let alone avoid stuff like binary compatibility breaks.
My objections are: (a.) most importantly the lack of any significant usable documentation, (b) very poor integration with VB6, you may as well be writing in some dark-world language like FreeBasic using a dumb text editor, (c.) too damned much in one giant bag, (d.) lack of a proper developer installation package with .DEP file(s), and (e.) how much misappropriated source might be contained within it.
Sure, there are open source licenses that allow usage with minimal attribution without disclosing your source, but those seldom apply when you are merely creating a wrapper around the code. In that case opening your source is a fair expectation of both users and the hijacked contributors.
-
Jan 22nd, 2018, 07:57 PM
#93
Re: Vb6 , the Future, and what I have discovered
@dilettante - thanks for the thoughtful post.
Originally Posted by dilettante
I worry about the "getting hit by a bus" scenario and the lack of future bug fixes that may follow, but not that much.
I'm in agreement with you there...
Originally Posted by dilettante
Even with the source code you might not have the necessary tools or expertise to recompile the thing.
And there.
Originally Posted by dilettante
My objections are: (a.) most importantly the lack of any significant usable documentation.
I've never been much of a "documentation-first" kind of programmer...partly because I've encountered some bad (and even downright misleading) documentation in the past, partly because the ability to step through code, watch variables, and modify "live" code features of VB reinforced my innate preference to learn by experimentation. That said, this issue has been brought up enough times by others that I think it needs to be addressed with proper documentation. Many years ago I offered to host a kind of documentation "wiki" for what was then dhRichClient, but nothing much came of it.
If Olaf is reading, this is an another "boring" area I'd be happy to contribute to.
Originally Posted by dilettante
(b) very poor integration with VB6, you may as well be writing in some dark-world language like FreeBasic using a dumb text editor
Sorry, this one I don't get at all? It integrates just fine as an ActiveX DLL. Intellisense works, and there are well named and demarcated classes. RC5 allows great integration with VB6 as far as I am concerned (If you know a better way to integrate Sqlite and Cairo into VB6, I'm all ears).
Can you provide examples of parts that are poorly integrated with VB6? Those areas would be prime candidates for future improvements.
Originally Posted by dilettante
(c.) too damned much in one giant bag
I'm torn on this one...on the one hand it seems to be a sticking point for a lot of potential adopters who think "it's too damned big!". On the other hand, the whole thing comes in at less than 10MB (I think?) which is really peanuts in this day and age. Another (paid) 3rd party library I use used to come as a whole bunch of individual DLLs. You;d have to make sure to package all the dependent DLLs if you used one small feature of one DLL that relied on another DLL (meaning frequent trips to the documentation if you have numerous little tools that use various DLLs). They recently(-ish - probably a couple of years ago now) moved to a single monolothic DLL. Some people still complain about that, but overall I find it much easier to deal with. Just copy and register a single file and use the parts you want to use without concern.
Originally Posted by dilettante
(d.) lack of a proper developer installation package with .DEP file(s), and (e.) how much misappropriated source might be contained within it.
Never used them, so I can't comment...I opt for your "favourite" InnoSetup, or pure registration-free, copy and run deployment instead of MSIs.
Originally Posted by dilettante
Sure, there are open source licenses that allow usage with minimal attribution without disclosing your source, but those seldom apply when you are merely creating a wrapper around the code. In that case opening your source is a fair expectation of both users and the hijacked contributors.
First, I am not a lawyer. That said, SQLite is public domain so the above is irrelevant. Cairo is dual LGPL/MPL. My understanding is that neither require you to distribute source for non-statically linked binaries. As much as we might like to see the source, I don't think there's any legal or moral obligation in this case.
Anyway, thanks for the civilized discussion points
Last edited by jpbro; Jan 22nd, 2018 at 10:21 PM.
-
Jan 22nd, 2018, 09:20 PM
#94
Fanatic Member
Re: Vb6 , the Future, and what I have discovered
Is there Any VB6 open source for SortedDictionary like .NET, in terms of performance?
-
Jan 22nd, 2018, 09:25 PM
#95
Re: Vb6 , the Future, and what I have discovered
My experience with documentation is that the coder who wrote the code is just about the worst possible (literate) person to document the code, but that presents an obvious problem for small projects cause....who else will write it?
I have a few programs with NO documentation, a fact that I often feel guilty about. I have another where I have well over 1000 pages of documentation, most of which is total garbage by now. The documentation was sometimes written after the code problem had been solved, sometimes as I was working out how to solve the code problem. As long as the I was reasonably right in my solution, and didn't come revisit the issue as I learned more about the problem (you never understand a problem until you have solved it the first time), then the documentation is pretty good. However, the project has evolved over years, so there are large blocks that are now totally incorrect, and even larger blocks that are mostly correct....but not quite. That sounds like what JBPro was saying. That documentation would be mostly right, and wrong in possibly subtle and essential ways.
It would be great to go back and extract the good part into final documentation. However, I did mention that the documentation is over 1000 pages, so extracting and editing is no small task.
Furthermore, I don't know whether the documentation is useful. I find myself revisiting some of it from different angles. So, there's a design document that discusses why it does what it does. Another document covers how it does what it does. A third kind of documentation that I'm just starting on is one that talks about what is there, and how it is used. All three contribute a certain amount to understanding, but no one of them is complete and all of them together is far from ideal.
My usual boring signature: Nothing
-
Jan 22nd, 2018, 09:27 PM
#96
Re: Vb6 , the Future, and what I have discovered
Originally Posted by DaveDavis
Is there Any VB6 open source for SortedDictionary like .NET, in terms of performance?
In my opinion, that should be a question in a new thread rather than one buried in here, but I'll leave it up to you to decide whether to start a thread on it, or not.
My usual boring signature: Nothing
-
Jan 22nd, 2018, 09:37 PM
#97
Fanatic Member
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Shaggy Hiker
In my opinion, that should be a question in a new thread rather than one buried in here, but I'll leave it up to you to decide whether to start a thread on it, or not.
I will open a new thread for VB6's SortedDictionary. Thanks for reminding.
In my opinion, VB6 is an old and obsolete language as Win95 vs Win10, which is 32 bit and even hardly to install on Win10. VB6 was previously famous with its fast development but comparing to .NET/JAVA, the knife is no more sharpening, for example, it lack of most of Array functionalities, lack of SortedList/SortedDictionary/Linq...
-
Jan 22nd, 2018, 09:40 PM
#98
Re: Vb6 , the Future, and what I have discovered
what's wrong with DarkBasic?
-
Jan 22nd, 2018, 10:13 PM
#99
Re: Vb6 , the Future, and what I have discovered
[QUOTE=Shaggy Hiker;5254865]My experience with documentation is that the coder who wrote the code is just about the worst possible (literate) person to document the code, but that presents an obvious problem for small projects cause....who else will write it?
Originally Posted by Shaggy Hiker
I have a few programs with NO documentation, a fact that I often feel guilty about.
No need to feel guilty if those programs are beings used well without documentation...how much software comes with documentation now (or if it does how much of it is being read)? Incredibly complex hardware devices with whole application ecosystems basically come with no documentation (smart phones). Now I know you are writing scientific software that is much more complex than your average angry birds game, but if it's being more or less intuitively understood by your user base, then I don't necessarily think you *need* documentation.
Originally Posted by Shaggy Hiker
over 1000 pages of documentation, most of which is total garbage by now.
The "total garbage by now" part is the hardest part of documentation - it's a continuous effort, especially for evolving project, but even a mature one will get all sorts of comments from new users that will make you want to update the documentation. For an individual programmer or a small group of programmers, this can be a killer. It's made worse by the fact that (in my experience) most users won't even read the document even if you have 100% perfect documentation included. "RTFM" is a commonly used abbreviation for a reason! Sometimes documentation is used defensively more than it is used preventatively.
Originally Posted by Shaggy Hiker
However, the project has evolved over years, so there are large blocks that are now totally incorrect, and even larger blocks that are mostly correct....but not quite. That sounds like what JBPro was saying. That documentation would be mostly right, and wrong in possibly subtle and essential ways.
More or less, yeah I haven't met much (if any) documentation that felt maliciously or deliberately wrong, but there have been a couple of MSDN pages that I've been looking at recently that made me wonder (and this is coming from someone who thinks that Microsoft's documentation is actually pretty decent, so not trying to start a fight here).
Originally Posted by Shaggy Hiker
It would be great to go back and extract the good part into final documentation. However, I did mention that the documentation is over 1000 pages, so extracting and editing is no small task.
There's a cost-benefit analysis here too...would you be better off making your software better/more intuitive, or writing more documentation against the current implementation...Sometimes I get feedback that make me move a button, or rename an element, or improve a workflow. That can have relatively huge implications on the documentation (new screenshots, updating text, re-numbering/re-ordering lists of steps, etc...). Sometimes those changes are cascading and can effect many areas/pages of documentation. Maybe that time is better spent elsewhere? Especially since (again, in my experience) many users don't read the documentation.
Originally Posted by Shaggy Hiker
Furthermore, I don't know whether the documentation is useful.
You and me both Or at least, I don't know if it's as useful as Dilentantte makes it out to be....It certainly wouldn't be the #1/Letter A) thing keeping me away from trying a new piece of software (TBH I can't remember the last time I sat down and actually read through any documentation...but I guess I'm a tinkerer/hack).
Originally Posted by Shaggy Hiker
I find myself revisiting some of it from different angles. So, there's a design document that discusses why it does what it does. Another document covers how it does what it does. A third kind of documentation that I'm just starting on is one that talks about what is there, and how it is used. All three contribute a certain amount to understanding, but no one of them is complete and all of them together is far from ideal.
One thing I have found that "works" (though I often have to send these steps in an email message even though they are in the documentation)....numbered lists of step by step (ala the Microsoft Manual of Style).
e.g.
1) On your Desktop, click the MyProgram Icon. The MyProgram window appears.
2) On the MyProgram window, click the File menu.
3) In the File menu click Open. The Select a File window appears.
4) Select the file that you want to open, and then click Open. Your file is now ready to be updated.
That kind of step-by-step stuff has always given me good results (and nice comments from users).
Last edited by jpbro; Jan 22nd, 2018 at 10:23 PM.
Reason: Can't believe I wrote "gotten" ;)
-
Jan 23rd, 2018, 03:51 PM
#100
Re: Vb6 , the Future, and what I have discovered
I was talking more about documentation for a person wanting to maintain/extend the software. Documentation for users? I HATE writing that. I've seen it done well, and I've seen what I have written, and nothing has ever been found in both categories.
My usual boring signature: Nothing
-
Jan 24th, 2018, 04:13 AM
#101
Re: Vb6 , the Future, and what I have discovered
Originally Posted by DaveDavis
Is there Any VB6 open source for SortedDictionary like .NET, in terms of performance?
Originally Posted by DaveDavis
I will open a new thread for VB6's SortedDictionary. Thanks for reminding.
In my opinion, VB6 is an old and obsolete language as Win95 vs Win10, which is 32 bit and even hardly to install on Win10. VB6 was previously famous with its fast development but comparing to .NET/JAVA, the knife is no more sharpening, for example, it lack of most of Array functionalities, lack of SortedList/SortedDictionary/Linq...
Perhaps the following information is useful to you:
http://www.vbforums.com/showthread.p...-concatenation
http://www.vbforums.com/showthread.p...=1#post5244857
VBCorLib (Its author seems to be @killian353535):
http://www.kellyethridge.com/vbcorlib/doc/VBCorLib.html
-
Jan 24th, 2018, 04:31 AM
#102
Re: Vb6 , the Future, and what I have discovered
IMO, compared with the document, most people are more willing to read "samples-code".
-
Jan 24th, 2018, 09:46 AM
#103
Re: Vb6 , the Future, and what I have discovered
Originally Posted by dreammanor
IMO, compared with the document, most people are more willing to read "samples-code".
Yes, these are called "Copy/Paste Cowboys." I suspect that a lot of people who want to plink away at coding have inadequate language and reading skills. This is as important as having fingers to type with, since as already pointed out it is pretty hard writing documentation so it places a burden on the reader.
In the early 1980s I worked for a large organization that ended up completely replacing "programming tests" by "reading skills tests" for programming jobs. The quality of hiring candidates improved immensely, with a higher rate of success. Then the "political correctness police" came along and dictated that any testing was discriminatory. Quality and rates of success plummeted from then on, and development costs began to skyrocket.
-
Jan 24th, 2018, 10:15 AM
#104
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Schmidt
A brand which has already proven "to not really care" about your code-investments
Didn't really wanna get involved in this shitstorm but this statement is patently false and highly misleading. MS devotes a tremendous amount of effort to backward compatibility. I remember reading an article some 7 to 10 years ago about the insane lengths MS was going to keep stuff working as they released new and newer versions of Windows. You can say a lot of bad things about MS but their commitment to backward compatibility is not one of them.
-
Jan 24th, 2018, 12:59 PM
#105
Re: Vb6 , the Future, and what I have discovered
When your ship is sinking fast with a roaring blaze consuming it you may say things you might regret later in the desperation of the moment. Doubly so if you pridefully refuse to get into the lifeboats while you try to fight the fire and bail the rising water at the same time.
-
Jan 24th, 2018, 02:02 PM
#106
Re: Vb6 , the Future, and what I have discovered
Originally Posted by dilettante
Yes, these are called "Copy/Paste Cowboys." I suspect that a lot of people who want to plink away at coding have inadequate language and reading skills. This is as important as having fingers to type with, since as already pointed out it is pretty hard writing documentation so it places a burden on the reader.
I don't completely agree with that... some of us learn best by doing rather than by reading. If I can pick up something, plink with it for a bit & figure it out, then I'm more likely to remember how to do it next time. Reading dry technical documentation doesn't really do that for me. That said, I'm not saying I never read the documentation, just that I'm more likely to quickly get up to speed on something by reading the sample code rather than the documentation. The T-SQL MERGE statement is a prime example. First few times I went to use it, I tried to read the documentation on it... after a couple minutes with that my mind was going into shut down mode. Skipped down to the samples... read through the sections there... found the ones closest to what I wanted... tinkered with it, re-read the relevant sections in the documentation I needed to modify... now, it's almost old hat for me. When I do need help with it, I'm more likely to read the text part of the entry as the samples are generic and basic.
Now, yes, I do agree there's a large amount of "Copy/Paste Cowboys" out there sadly -- I call them Code Slingers... they're the ones slinging code about without truly understanding what it does or how it works. Those are the ones that are dangerous.
-tg
-
Jan 24th, 2018, 02:17 PM
#107
Re: Vb6 , the Future, and what I have discovered
A brand which has already proven "to not really care" about your code-investments...
Originally Posted by Niya
... this statement is patently false and highly misleading.
It is not - for the following reasons:
- I wrote "has proven" -> past tense (at the time they abandoned VB6, there were absolutely *no* statements for years, that they intended to support at least the runtimes till 2024)
- abandoning (deprecating) an entire language without a viable upgrade-path was *devaluing* millions of man-years of development-efforts across the globe (unheard of, on that scale)
- I have personally had my self questioned, which language my solution was based on, and anwering truthfully had the effect that I had to underprice my products to remain "in the game"
- "arguing" that VB6-solutions will "probably work for a few years still" (because of MS's "huge efforts with backward-compatibilitiy") would only have caused a "mild smile" on the side of these customers
So, no - I stand by what I wrote above.
Professional Software-Development is for the most part "investment of time" (a highly priced resource, because the people who invest that time professionally, are educated, highly skilled ones).
And these code- (and thus time-)investments will pay off only in the long-run (we often talk decades here, not weeks or months as in other branches).
A Vendor who provides a commercially sold language, which is the base for all these huge investments, has a responsibility which is also not comparable to "other industries", end of story.
(especially when we talk about the "most successful language on the planet" (at that time having about 6Mio users).
Simple calculation of what I mean? - here we go (incorporating only 10% of these 6Mio developers as professionals - and thus "most affected by the abandoning"):
Invested man-years of these 600,000 Pros (assuming 3 years of "working hard" on a product on average):
600,000 professionals * 3 man-years = 1,800,000 professional man-years invested
Monetary value of the above calculated 1.8Mio man-years (assuming a real conservative average of $50,000 per man-year):
1.8Mio man-years * $50,000 per man-year = $90,000,000,000 (90 Billion Dollars)
Expected ROI for these invested "monetary expressed man-year-efforts" - perhaps with a marge of 30%?
90 Billion Dollars * 1.3 = 117 Billion Dollars.
Now we could all "argue felicitously" about the "average rate" that expected "total ROI" was devalued
(by either not being able to sell ones VB6-solution(s) anymore, or by having to offer prices with next to no profit-margin).
But the resulting sum which was "lost" (to "us") is still in the two-digit-Billions!
Any insurance-company (or bank) who'd have tried a stunt in that dimension (devaluing the investments of their customers in that magnitude - and without need!)
would have seen a "sh..t-storm" with no end.
Aside from that, "lost trust" is a personal (subjective) thing - no real need to argue about it - but at least I've tried to back my position up with a few numbers.
Olaf
Last edited by Schmidt; Jan 24th, 2018 at 02:27 PM.
-
Jan 24th, 2018, 02:54 PM
#108
Re: Vb6 , the Future, and what I have discovered
That's only an issue because everybody was using MS. They've done better than any other, but most others can make a breaking change and will offend few people simply because they have few users.
Still, when you are tied to a company, or tied to a technology, then you stand on unstable ground.
My usual boring signature: Nothing
-
Jan 24th, 2018, 03:19 PM
#109
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Shaggy Hiker
That's only an issue because everybody was using MS.
They've done better than any other, but most others can make a breaking change and will offend few people simply because they have few users.
A "breaking change" (as happens e.g. in a given "library-stack" from time to time) is entirely normal (we had a taste of that with VB3-VB4-VB5 transitions) -
and no larger problem, because that's stuff, you usually can fix ("refactor, or re-route" to new, quite similar and often mightier interfaces) within a few hours, days( or in rarer cases weeks).
Abandoning an entire language is a completely different thing (time-investment-wise, because it affects all your tested code with perhaps hundreds of thousands of lines).
Did you ever hear of something on that scale happen (for big, widely and professionally used "non-experimental" languages)?
- C/C++?
- Delphi/ObjectPascal?
- Cobol?
- Fortran?
- Visual Objects?
I deliberately listed those with a larger user-base, and for which "commercially sold compilers and environments exist" -
(leaving out Java, JavaScript, Python, PHP etc.)
Olaf
Last edited by Schmidt; Jan 24th, 2018 at 03:22 PM.
-
Jan 24th, 2018, 04:40 PM
#110
Re: Vb6 , the Future, and what I have discovered
C/C++: ANSI Standard languages. So, yes, there were such changes, but only in the run up to the first ANSI standard ratification for each of them. That means back before my time for C, but I was watching the contortions around the standardization of C++ back in the early 90s. If you bet on the wrong horse, you lost.
Delphi: Somebody listed serious breaking changes in this language fairly recently over in General Developer. I forget who it was, but I seem to remember that it was a moderator. Apparently, they were willing to break all existing code practically on a whim. If you wanted to hunt for it, I think the thread was started by axisdj when he was considering a future path.
No idea about the others.
My usual boring signature: Nothing
-
Jan 24th, 2018, 06:06 PM
#111
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Shaggy Hiker
C/C++: ANSI Standard languages. So, yes, there were such changes, but only in the run up to the first ANSI standard ratification for each of them. That means back before my time for C, but I was watching the contortions around the standardization of C++ back in the early 90s. If you bet on the wrong horse, you lost.
I was asking sepcifically for vendors who had cancelled their language.
That there might have been "larger differences" e.g. between "Borland C/C++" or "MS C/C++" or "Intel C/C++" (especially in the early days before ANSI standardization) is without question -
but the (perhaps more incompatible in the beginning) compilers were not abandoned by their vendors.
Originally Posted by Shaggy Hiker
Delphi: Somebody listed serious breaking changes in this language fairly recently over in General Developer. I forget who it was, but I seem to remember that it was a moderator.
Apparently, they were willing to break all existing code practically on a whim. If you wanted to hunt for it, I think the thread was started by axisdj when he was considering a future path.
I've just googled directly (in Delphi-Blogs, where developers described their migration-experiences -
e.g. for the Unicode-Changes in the language and certain older Control-Stacks which took place recently) -
and from what the bloggers wrote, it seems there were no larger problems, nothing which came even near to "abandoning the language itself"...
For every major (new) feature I've found, Embarcaredo seems to have offered Migration-assistents that apparently actually worked -
quite similar to what VB4 was doing, when importing from the older VB3 *.vbx'es and stuff - business as usual with "library-changes" (taking a week or two at most).
And I'd certainly like to hear such stuff first-hand, from someone who has actually used Delphi over the last years (perhaps someone is listening and helps out)...
Olaf
-
Jan 25th, 2018, 10:03 AM
#112
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Schmidt
It is not - for the following reasons:
- I wrote "has proven" -> past tense (at the time they abandoned VB6, there were absolutely *no* statements for years, that they intended to support at least the runtimes till 2024)
- abandoning (deprecating) an entire language without a viable upgrade-path was *devaluing* millions of man-years of development-efforts across the globe (unheard of, on that scale)
- I have personally had my self questioned, which language my solution was based on, and anwering truthfully had the effect that I had to underprice my products to remain "in the game"
- "arguing" that VB6-solutions will "probably work for a few years still" (because of MS's "huge efforts with backward-compatibilitiy") would only have caused a "mild smile" on the side of these customers
So, no - I stand by what I wrote above.
Professional Software-Development is for the most part "investment of time" (a highly priced resource, because the people who invest that time professionally, are educated, highly skilled ones).
And these code- (and thus time-)investments will pay off only in the long-run (we often talk decades here, not weeks or months as in other branches).
A Vendor who provides a commercially sold language, which is the base for all these huge investments, has a responsibility which is also not comparable to "other industries", end of story.
(especially when we talk about the "most successful language on the planet" (at that time having about 6Mio users).
Simple calculation of what I mean? - here we go (incorporating only 10% of these 6Mio developers as professionals - and thus "most affected by the abandoning"):
Invested man-years of these 600,000 Pros (assuming 3 years of "working hard" on a product on average):
600,000 professionals * 3 man-years = 1,800,000 professional man-years invested
Monetary value of the above calculated 1.8Mio man-years (assuming a real conservative average of $50,000 per man-year):
1.8Mio man-years * $50,000 per man-year = $90,000,000,000 (90 Billion Dollars)
Expected ROI for these invested "monetary expressed man-year-efforts" - perhaps with a marge of 30%?
90 Billion Dollars * 1.3 = 117 Billion Dollars.
Now we could all "argue felicitously" about the "average rate" that expected "total ROI" was devalued
(by either not being able to sell ones VB6-solution(s) anymore, or by having to offer prices with next to no profit-margin).
But the resulting sum which was "lost" (to "us") is still in the two-digit-Billions!
Any insurance-company (or bank) who'd have tried a stunt in that dimension (devaluing the investments of their customers in that magnitude - and without need!)
would have seen a "sh..t-storm" with no end.
Aside from that, "lost trust" is a personal (subjective) thing - no real need to argue about it - but at least I've tried to back my position up with a few numbers.
Olaf
All this is moot since VB6 applications and the VB6 IDE still work fine to this day, aside from some issues with installation and security. So all those countless man hours spent developing VB6 applications hasn't been lost. I think MS was more than fair with the amount of time they gave VB6 developers to adapt to the changes they made to the development ecosystem. Now if you spent the last 20 years complaining instead of adapting and they completely kill support for VB6 past 2024 then that's on you. You can't blame MS for this.
-
Jan 25th, 2018, 10:25 AM
#113
Re: Vb6 , the Future, and what I have discovered
Yeah, use VB6 if you choose to as long as the ride lasts. If VB6 is no longer viable down the road, well... then you'll have to make choices.
There isn't any obvious backup plan though. Anything else out there is even more likely to mutate and break programs or disappear anyway. Life is change.
-
Jan 25th, 2018, 11:34 AM
#114
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Niya
All this is moot since VB6 applications and the VB6 IDE still work fine to this day, aside from some issues with installation and security. So all those countless man hours spent developing VB6 applications hasn't been lost. I think MS was more than fair with the amount of time they gave VB6 developers to adapt to the changes they made to the development ecosystem. Now if you spent the last 20 years complaining instead of adapting and they completely kill support for VB6 past 2024 then that's on you. You can't blame MS for this.
Hi Niya,
You are correct in the first part of what you say, thank goodness!
However, and I hate to say it because Olaf and I tend to go round-and-round at times, but I must agree with him. Microsoft screwed us over royally with the move to .NET. And I still insist that there was no call for it, other than someone's ego at Microsoft. I have upgraded to different compilers and IDEs many times through the years, and everything always went "fairly" smoothly until the attempt to jump to .NET.
I remember quite well the loss of the Field statement and having to rework all my ISAM file accessing methods. It was a big argh at first, but once I got going, it all went fairly smoothly with UDTs and the Open Random method.
However, attempting to translate a major and well developed application from VB6 to .NET was (and still is as far as I understanding) a virtual rewrite. I remember taking a run at it with a couple of my applications, and I always just threw my hands up, thinking I had better things to do with my time.
And I feel certain that I'm far from the only one who hit this brick wall. Why else would this particular forum be as active as it is.
Microsoft truly screwed us with respect to VB6. And, worst of all, there was really no call for it. Why not just include both the COM and .NET architecture in their more recent compilers? You can't say that they wanted to abandon all COM programming as that's still what all their VBA programs/macros use. I don't see how this is anything but pure ego on someone's part at Microsoft, back when they introduced .NET.
Well, back to work, plugging along with my old COM programming.
It'd be cool if someone disassembled the VB6 IDE, spent a bit of time commenting a few of the sections to make it minimally interpretable, and then anonymously posted it. If even a remotely interpretable version of it got into the wild, I truly believe people would go nuts with it. But hey, I'm not suggesting that (he says for legal reasons).
Y'all Take Care,
Elroy
EDIT1: Just now checked:
A 20 year old (non-updated) language that does a respectable job of keeping up with the latest Microsoft language is actually quite impressive.
Last edited by Elroy; Jan 25th, 2018 at 11:44 AM.
Any software I post in these forums written by me is provided "AS IS" without warranty of any kind, expressed or implied, and permission is hereby granted, free of charge and without restriction, to any person obtaining a copy. To all, peace and happiness.
-
Jan 25th, 2018, 12:09 PM
#115
Re: Vb6 , the Future, and what I have discovered
Hi Elroy,
I also missed the Field-Statment
Originally Posted by Elroy
I remember quite well the loss of the Field statement and having to rework all my ISAM file accessing methods. It was a big argh at first, but once I got going, it all went fairly smoothly with UDTs and the Open Random method.
but it isn't all that much diffrent...
Code:
Public Shared Function Read(ByVal ID As Long) As clsAdressen
Dim Dr As DataRow
Dim sSQL As String
sSQL = "SELECT AN_ID,"
sSQL = sSQL & "AN_Vorname,"
sSQL = sSQL & "AN_Nachname,"
sSQL = sSQL & "AN_Angelegt"
sSQL = sSQL & " FROM tbl_Adressen"
sSQL = sSQL & " WHERE AN_ID=" & ID
Dr = clsDBConnection.GetRowInfo(sSQL)
Read = New clsAdressen
With Read
.ID = CType(Dr(0), Long)
.AN_Vorname = CType(Dr(1), String)
.AN_Nachname = CType(Dr(2), String)
.An_Angelegt = CType(Dr(3), Date)
End With
End Function
the other alternative is to wright a Class in . Net that handles then -Field Statment- as you know it
regards
Chris
to hunt a species to extinction is not logical !
since 2010 the number of Tigers are rising again in 2016 - 3900 were counted. with Baby Callas it's 3901, my wife and I had 2-3 months the privilege of raising a Baby Tiger.
-
Jan 25th, 2018, 12:19 PM
#116
Lively Member
Re: Vb6 , the Future, and what I have discovered
@Elroi
... well your right, but I don't think it was just a big-ego at MS, but a hard core business decission.
Originally Posted by Elroy
A 20 year old (non-updated) language that does a respectable job of keeping up with the latest Microsoft language is actually quite impressive.
I think, they just suspected exactly that!
I mean, how would these numbers look, if they did'nt kill VB6?
-
Jan 25th, 2018, 08:00 PM
#117
Re: Vb6 , the Future, and what I have discovered
Hi tubus,
How would those numbers look? I suspect they would have recognized the error of their ways, and the COM architecture would still be a frequently used platform for new software developments. I mean, I won't deny that .NET has its advantages, but why does it have to be one-or-the-other? Why not both?
And, so what if people kept (keep) using COM? I suspect that Microsoft might be doing better in markets such as tablet, phones, phablets, and the like. I've advocated for years that they should build a p-code interpreter right into Windows. Earlier predecessor OSs we based on BASIC (including the first OS by Microsoft), so why not? They'd have every nerdy kid in the world fascinated with writing little apps for all the Windows devices. IMHO, Microsoft totally blew it.
Also, IMHO, it's really not too late to "fix" it. Still, I believe, if they were to just release a COM based p-code interpreter (much like the VB6 IDE Learning Edition) that was built into Windows, people would think it was better than sliced bread.
I don't see what possible business decision would prevent Microsoft from doing this. That's why I say "ego". To me, there's just no other explanation.
@ChrisE: Yeah, I left the Field statement behind about, erm, 23 years ago. I was just trying to point out that all BASIC language transitions weren't that significant until .NET was introduced.
All The Best,
Elroy
Any software I post in these forums written by me is provided "AS IS" without warranty of any kind, expressed or implied, and permission is hereby granted, free of charge and without restriction, to any person obtaining a copy. To all, peace and happiness.
-
Jan 25th, 2018, 08:21 PM
#118
Re: Vb6 , the Future, and what I have discovered
COM and OLE2 (ActiveX) are still native technologies, and probably always will be. Both Shell32 and MS Office (just for starters) rely on them extensively. And UWP (WinRT) is based on COM extensions.
-
Jan 26th, 2018, 05:53 AM
#119
Lively Member
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Elroy
How would those numbers look? I suspect they would have recognized the error of their ways, and the COM architecture would still be a frequently used platform for new software developments. I mean, I won't deny that .NET has its advantages, but why does it have to be one-or-the-other? Why not both?
...I suspect, these numbers would look like 10 times more in favor of VB6!
And if they took the parallel road, it would hav been obvious to everybody, how lame in comparison VB.NET (at the beginning at least) was.
But i'm with you, was as angry as you with the way they cheated on its legit customers.
Not so much for the facts itself, but first of all, that they did'nt tell the truth.
All that talk about easy migration, or in the words of one of its reps: selling the public a good upgrade story! Bullshit!!
And I really tried for quite some time, until I realized it was close to one thing: mission impossible!!
..
-
Jan 26th, 2018, 06:08 AM
#120
Lively Member
Re: Vb6 , the Future, and what I have discovered
Originally Posted by Elroy
... I suspect that Microsoft might be doing better in markets such as tablet, phones, phablets, and the like.
..yeah, I think this too. Mainly because VB6 apps can be so incredibly compact.
But, alas, tablets, phones & the like means multitouch
And how to do that in VB6??
May be like this:
(I'm far from beeing an expert in things like this and have little knowledge if this could be viable)
with a layer in the system, which translates multiple touches into a series of mouse_down and mouse_move events in a round robin way?
For several type of apps this might work (thinking of so called control surfaces for audio devices/programms with sliders & knobs & the like.)
One can dream
..
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules
|
Click Here to Expand Forum to Full Width
|