Replies: 5 comments 21 replies
-
|
What in particular is invisible? Both features involved in the announcement are sourced from the fslang-suggestions repo: fsharp/fslang-suggestions#1262 And the first is precisely a case of community involvement, wherein something originally closed as "won't do" was argued for openly and eventually accepted into the language. |
Beta Was this translation helpful? Give feedback.
-
|
I also find it difficult to see what's currently happening with F#, since there are a lot of RFCs (some pretty old) and almost all the updates I get are about formatting/style. I second the idea of some landing-page showing what is currently being worked on. Maybe I'm just missing something. |
Beta Was this translation helpful? Give feedback.
-
This is pretty much just issues and PRs in this repo. It shows in detail what's happening here. Project is also good source of what's team is planning (though it's oftentimes randomized with current PR reviews, bugfixes and some additional work, which is reflected in this repo activity too). If more detailed info is desired, I personally will not be opposed to FSSF maintaining such separate page based on said sources. |
Beta Was this translation helpful? Give feedback.
-
That sounds interesting @vzarytovskii, and I had something similar in mind, too. Could be a blog with small summaries / currated links of the current state of F# language / core / tooling progression; with a narrow focus on exactly these things. Could be done via FSSF, but also a community thing in addition to Sergey's F# Weekly (which is my go-to place and source for general F# and related infos), and Amplifying F# (as a hands-on support for getting-things-done in the compiler and ecosystem). The goal(s) of such a thing could be
Maybe I'll give it a try... Thanks for all your comments (more appreciated if you think it's a thing worth to try). |
Beta Was this translation helpful? Give feedback.
-
|
I understand the desire to have a middle-ground between waiting for an official "F# vNext just released" blog post and actively watching the dotnet/fsharp repo. That being said, I also have to emphasize that chances and timelines for a PR in progress are not predictable at all. I have seen the wave of negative feedback for "Tentative C# 12 feature list" and was very disappointed by it. Arousing interest is good, frustration when items do not eventually make it - not so much. If anyone wants to curate a list of finished and WIP pull requests that relate to a language change, I think I can offer marking them with a specific label whenever I come across them ("e.g. 'Area-Language-Change'). That way, a URL with specified GH filter can be used to irregularly watch them without having to go trough all the other PRs (bugfixes, bots, tooling changes,...). |
Beta Was this translation helpful? Give feedback.
-
|
@T-Gro Unlike C# language design, here in F# land if something is not on the plans, we can easily invite those dissatisfied to suggest, write RFC or even go for an implementation. C# design is heavily gated by the team from Microsoft while F# design is much more open and up-for-grabs. I think we will be in a better shape than C# here. |
Beta Was this translation helpful? Give feedback.
-
|
Follow-up change: dotnet/fsharp#15901 |
Beta Was this translation helpful? Give feedback.
-
|
It's not ideal, but these days there's a pretty canonical list of new language features at https://github.com/dotnet/fsharp/blob/0bfd490b47fd35c4a48c88a1b2fd8c50b724f3e1/src/Compiler/Facilities/LanguageFeatures.fs#L113 . The two new features mentioned above are |
Beta Was this translation helpful? Give feedback.
-
Yes, that's sad. When I talk to my clients (mostly C# / TS / whatever-average-client-stuff), they don't even see F#, or have heared about it, but don't really understand what's going on or if it's still "alive". Part of the reality - like it or not - is this: The louder one shouts, the more often one does it, the more visible one becomes, where visibility is at least a factor in becoming or staying relevant. There's a proverb, that goes like this: "As a politician, it's better to have bad publicity than none". Now I'm not saying that it's good to spam people, shout at them or so - and it's not necessary - because there's so much good stuff here! Instead of having no or bad publicity, we can have good one! In the end, there might be less publicity than we could have. Personally, I'm not good at marketing, and I have no clue how to do it, and I love spending my free evenings developing software, rather than doing a marketing campaign or something. But in the end, I'm just convinced that F# is one of the most practical languages out there, and I'm happy to see that so many people put effort in it, with enthusiasm, and that's something that's unique - at least for the things I know. Hopefully, I'll find a way, time, mood and spirit for making an editorial format that will make everything that's going on in this great community more visible; at least a little bit. SOmething about the language progression, the community, the people. Failure is always an option; maybe I won't be able or it will just be crap but I will try... |
Beta Was this translation helpful? Give feedback.
-
Main reason is that we don't have technical writers on the team, and people are usually busy with tasks they're working on, so release notes are not simply sent to editorial team. |
Beta Was this translation helpful? Give feedback.
-
Yes, I can understand that absolutely. Devs are devs - not PR or marketing people. It's a very different role. It's not something that runs under the hood; it's a dedicated field. Hopefully, there's a way of more or less quickly go through various sources like dotnet/fsharp, fslang-X repos, Sergeys blog and others, and then distill some concise articles. |
Beta Was this translation helpful? Give feedback.
-
|
Good recent places to start on what's been happening would be the first and last talks of the recent fsharpconf, which can be found on the dotnet YouTube channel. The first is Kathleen and a few friends, the last is the F# team talking about features and stuff. |
Beta Was this translation helpful? Give feedback.
-
|
The one with Kathleen, Vlad and you - yeah I got that in view. The other one I guess not, but I'll find. Thanks got the hint. |
Beta Was this translation helpful? Give feedback.
-
|
I'm working on an idea and attempt for making F# language and compiler progression more prominent, currently by playing with web-scraping fslang-suggestions and dotnet/fsharp repos + manuel editing to render some small pages, giving an overview of the current state and their transitions. I thought that this might be possible with a minimal effort in order to keep such a thing alive, but still providing value to the viewer. It's a very rough idea + fiddling around. Roughly, it might come out to something like: A thing that could be "done" (in the past or fufute) is the last element of a stream of events, made up of various non-formally coupled sources.
An example: fsharp/fslang-suggestions#1262
Is there a way to understand if work started on a fslang-suggestion of a (bigger) feature in dotnet/fsharp? Just from the suggestions, it seems like it's suddenly done (I mean: That's an information - and a great one indeed :) ) Pls excuse if my language might sound a little bit "clumsy"; I tried my best. And thank you for your support. |
Beta Was this translation helpful? Give feedback.
-
|
The screenshot is actually a very good example. As soon as a PR is raised (and that is very often a starting work in progress, just a draft), the original suggestion is being linked. The linkage is there while the PR is still open, independent of being later merged or not. This tells you that the work has been started. |
Beta Was this translation helpful? Give feedback.
-
|
The label "Merged" on the right side is automatically derived by GH. Before it was finished, the linkage was there as well, just the label would have been green and saying "Open". |
Beta Was this translation helpful? Give feedback.
-
|
Side note, the title can be fixed to |
Beta Was this translation helpful? Give feedback.
-
|
I kinda liked the |
Beta Was this translation helpful? Give feedback.
-
As briefly discussed on https://twitter.com/dawe70704856/status/1696430171882975604, my personal feeling is that transparency in the development of F# has decreased lately. Maybe that feeling is wrong, but I can give some examples representative for it that I hope might clarify what this is initially about. My very general expectation of a major open source project which is open for partifipation and contribution regarding transparency is:
Surely there are other points and you don't have to share the points mentioned. That F# is not a pure community project, but a largely Microsoft funded and directed investment (thanks for that; no irony here), is clear to me. I'm also not concerned with democratizing decision-making processes here. Even the way decisions come about doesn't necessarily have to be open or well-reasoned (to me). What I think is necessary, however, is "what" is happening or will happen. Why's that important, actually? A subjective answer: for me it was and is always a great experience to get in touch with people from "the community". F# is not a huge thing, but I always had the impression that for many, a very specific feeling is of importance, which goes beyond the (in every respect remarkable) language F#. To have the feeling of "something is moving and I can make a difference": C#, TypeScript - they don't need that. F#, perhaps, does. For me it was always of great importance; I liked it and could write a lot about it! I simply connect a bunch of good feelings and memories with F#; as crazy as it sounds: That's the truth.
From the README of this repo:
So transparency and participation are intended - I had no doubt about that. So it's more about what it means or how it can be achieved. By the way: As a passive part of the F# language and tooling development I only refer to transparency - not to active participation here.
What has changed concretely from my point of view: I used to focus on fslang-suggestions and fslang-design. While the former gave (and still does) a rough sense of what might happen and why, fslang-design was the place to understand what will most likely happen or is currently happening. But also what has happened - so important as a de facto place of prose specification and explanation. That was enough for me to think I was "well informed". This didn't require F# Discord, Slack, Twitter or so. Meanwhile, I have the impression that important things no longer catch on. Maybe I simply do not know where to look - can that be?
Again: This is a very subjective account. If that opinion shared, I could imagine a few simple things that could lead to more transparency:
Bottom line, I just think it's sad when hard work (from everyone who participates) is so invisible. Thanks to everyone - internal and external, paid or unpaid, pro or amateur - for the great commitment, which also benefits developers like me.
Have a nice day.
Beta Was this translation helpful? Give feedback.
All reactions