dotnet / vblang

The home for design of the Visual Basic .NET programming language and runtime library.
290 stars 66 forks source link

MS should help VB community to take over! #445

Open VBAndCs opened 5 years ago

VBAndCs commented 5 years ago

I am under the impression that @AnthonyDGreen is working on VB.NET alone! Even this repo seems nearly neglected due too few interactions with the team / admins. If MS doesn't want to waste its resources on VB.NET and let it for the community, this is OK, but we have an essential problem: VB.NET programmers that have a deep knowledge of compilers at general, and VB Compiler/Roslyn SDK in particular, are rare (VB6/VB.NET was not a first choice for compiler builders), so, the community can't easily add new features to VB languages and code editor. To change that, MS should provide us with all resources needed to learn the VB compiler: documentations, notes, tutorials, videos, and even private tools that MS use to build and test VB new features. If you want to give up VB, fine, but hand off the control to the community. Publish every thing in details in a wiki in this repo, provide PDF books, and let us take care of our beloved language. VB community needs a strong push under some supervision at beginning. Create a VB.NET lab to test suggested features, assign one of the team to it, document every detail about how features are added, and assign some work to persons willing to learn and participate. I don't think there is an easy alternatives. There is no other place to teach us such things. I am not aware of MS certificate of this topic. If someone decided to take this journey alone, it will take years with lack of resources and assistance. So, Ms should help to build this community from ground up, otherwise, we will continue proposing ideas with no one or no mean to implement them.

DualBrain commented 5 years ago

Although I agree in underlying sentiment... I disagree with the approach. This isn't something that is "Microsoft's responsibility". It's ours. I, for example, am already in the process of working on a few things... where these are in a small holding pattern due to the flux that is .NET 3 to .NET 5 timeframe. However, with that said, there are already several opportunities to help rebuild community... however, it takes community to build community. I don't believe Microsoft is opposed to this. With all that said, I also don't think that "community" will really be too involved in the actual hands-on work with regards to the compiler (Roslyn). However, tons of opportunity to get involved with additional tools, libraries, documentation, etc. And, ultimately, without the involvement in these areas already established... what's the point with "community" being involved at the compiler level if there isn't anyone in community able to test, document, evaluate, etc. Additionally, there is so much available with regards to what can already be done that there are so many open opportunities to be involved on already existing things... there are still a ton of developers that are still "new" (to one degree or another) and don't have deep details with regards to existing technologies... not to mention work-arounds... which, regardless of how much work is done... this will always be necessary. I recently saw a conversation somewhere there the topic was on how much information (surprisingly) has disappeared from around 2005... much of which is still absolutely relevant and possibly even more so with the revamp that is happening with WinForms on .NET Core. Yet another conversation I saw recently was asking about topics at events... mix of current and new seems to be what is desired and the message was clear that not enough "what you can do today" is being discussed. It's this constant chase to "what's next". Many people want to know "what now". Next is interesting, dare I say "sexy". However, current can also be cool. And, as it turns out... needed. My point is there is a lot that the VB community can do today that doesn't require Microsoft to do anything (other than to support - or rather encourage - where it makes sense) and we really need to start with the broader picture... where, maybe... and I do mean maybe... we approach Roslyn. It makes no sense to do it in the reverse...

Berrysoft commented 5 years ago

Of course it's our responsibility, but it is obvious that Microsoft is paying more attention on C# and F#. The LDM of Visual Basic doesn't have meetings that often as the LDM of C#, and they seldom approve new features, even the feature is only asking for the same ability as C#, for example, using of ref structs - they say it is unsafe, but Span<T> is aimed to make the operation fast and safe.

zspitz commented 5 years ago

@Berrysoft But there isn't very much Microsoft can do without active community involvement. That was originally the case for C# and .NET in general, until Omnisharp blew open the doors for writing C# in any editor, not just Visual Studio; it provided the strongest initial push towards .NET Core.

I think it falls to the community to be actively involved. Filing issues on proposed language features is just the beginning -- we as members of the community who feel passionately about this must write specifications, propose implementations, test proposed implementations, and write documentation.

It is also important for us to understand that even if the effort we put in doesn't make it all the way to the next version of VB.NET (in all probability, our efforts will not make it), the effort is not wasted, because it serves as a springboard for further discussion of the final version. It's much easier to point to specific lines in a PR and say "this will cause performance degradation", "what about in that edge case"; than to try and cover all bases from the beginning.

VBAndCs commented 4 years ago

I never reacted to this because no one of the team cared to respond! I early got the feeling that we are shouting in the void (/sub in VB :D). They were determinate to sink VB.NET ship and let it go! We were not the problem in first place!

zspitz commented 4 years ago

In light of Anthony D Green's blog post, I have to reconsider. All the community contribution requires a small 3-5 member team (preferably within Microsoft, but perhaps even outside) devoted entirely to the advancement of VB and its ecosystem. This team would be accountable for coordinating community contributions, such as templates, samples, compiler PRs; and would have the authority to inform/enforce changes on other teams within Microsoft when required by VB (within reason of course).