Add features matrix to the README #1593
Conversation
|
Would really love peoples opinion on this. Did I miss anything important? Any points I make do you think don't matter? |
Codecov Report
@@ Coverage Diff @@
## master #1593 +/- ##
==========================================
- Coverage 75.25% 74.99% -0.27%
==========================================
Files 78 78
Lines 5610 5610
==========================================
- Hits 4222 4207 -15
- Misses 1020 1034 +14
- Partials 368 369 +1
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
|
@albrow @backkem @adwpc @AeroNotix @Antonito @ashellunts @at-wat @audiolion @baiyufei @billylindeman @bshi @cgojin @cloudwebrtc @cryptix @decanus @enobufs @epes @hugoArregui @jeremija @jech @leewardbound @lherman-cs @masterada @maxhawkins @mission-liao @OrlandoCo @r-novel @rviscarra @RyanGordon @saljam @scorpionknifes @seppo0010 @sgotti @steebchen @TannerGabriel @tarrencev @tekig @thinkski @trivigy @wdouglass Would this have improved/detracted from your first opinion of using Pion? I have gotten a lot of feedback that people have no idea what |
|
https://github.com/pion/webrtc/blob/eef6045bb0203f0223f809cf0187571297b6adf6/README.md is a link directly to the file to see it formatted. |
README.md
Outdated
| * Easy integration with x264, libvpx, GStreamer and ffmpeg. | ||
| * [Simulcast](https://github.com/pion/webrtc/tree/master/examples/simulcast) | ||
| * [SVC](https://github.com/pion/rtp/blob/master/codecs/vp9_packet.go#L138) | ||
| * [NACK](https://github.com/pion/interceptor/issues/7) |
There was a problem hiding this comment.
why there is a link to an issue? it means not implemented?
There was a problem hiding this comment.
Almost there! It is in a PR here but not merged yet.
Bandwidth Estimation is in Ion and I am hoping we can just move into interceptors soon!
Forward Error Correction isn't available at all yet :(
Do you think it is better to leave these out? I will update the NACK link, but was going to link to those issues to track them until complete. I also thought about maybe putting symbols next to them to mark them as 'In Progress'
There was a problem hiding this comment.
Assuming the answer is yes (sorry, going to bed soon 😂 ), I don't think it is really clear (even if the answer is no I think it's not clear!). Also: I think it's easy to get outdated links once we close or deprecate issues. I think maybe for TODOs we can have a link at the bottom with a big global issue pointing to the current work, or maybe even to an issue search with label next-feature or something. Just my 2c!
|
Would not have changed anything for me. WebRTC is a fairly complex tech stack and usually requires decent know how. By the time I was using it full force I already knew what is inside. P.s. but the extra docs are always very welcome! |
README.md
Outdated
| * Plan-B and Unified Plan | ||
| * [SettingEngine](https://pkg.go.dev/github.com/pion/webrtc/v3#SettingEngine) for Pion specific extensions | ||
|
|
||
| #### Pure Go |
There was a problem hiding this comment.
minor feedback: this section in between the webrtc specific functionality seems a bit out of place
|
Definitely good improvement of README! |
|
I like this! My two cents: the On the other hand, the |
|
Looks good to me! |
|
It seems like a great improvement! |
|
Love this change 🙌 |
|
With all due respect, Sean, I think the "feature matrix" is overselling Pion. It might raise people's expectations too much, and then cause them to turn away from Pion when they realise that we've been overselling. I recommend marking Simulcast and SVC as "work in progress". I don't know what is the status of mediadevices, please mark it as experimental or something. Please be clear upfront that the user must bring their own implementations of congestion control and loss recovery. |
Sean-Der
force-pushed
the
issue-1279
branch
3 times, most recently
from
fe9f788
to
0ec266f
Compare
We split every protocol out to encourage them to be used on their own. Some didn't get much attention while others, like DTLS, grew their own vibrant community. It just happened naturally. Maybe improving the readmes of those other libraries may help attract eyes thought. |
|
Many complex terms can scare people off or raise their expectations. They may say "A lot has been done here, we can easily use this project", but they are not. I think that people who understand will not switch to another project, because they already have a ready-made solution that works. Not sure, but people come here who don't know much about WebRTC in the beginning. The words ICE, TURN and etc wouldn't give me anything. I wanted to see an example that works and is simple so that a quick start can be done. I think at the beginning you need to put a quick start, only then the features. An example should implement the most popular usecase, and also offer examples for less popular usecase. These examples should work in the real world so that the user does not think they have been fooled. Sorry a little off topic. |
|
@tekig what do you think of pushing features down lower? We can have I think we do have some people coming to this and evaluating how polished Pion is (where I think this list helps). I think you are right though about the majority of users don't care about feature set and just want to get going. |
Having this list of features is great, although the README is starting to get busy -- maybe better to have a 'Features' page on the wiki or pion.ly and link to it from the README.md? Personally, I learn best by example, so the example applications/demos were the most useful to get a feel for Pion. |
a good idea |
List most popular features/benefits of Pion WebRTC. Resolves #1279 Co-authored-by: ZHENK <[email protected]>
|
I'm late on this, but it's a good change! |
List most popular features/benefits of Pion WebRTC.
Resolves #1279
Co-authored-by: ZHENK [email protected]