Skip to content

Commit

Permalink
update with tldr of previous google doc and more on untraced
Browse files Browse the repository at this point in the history
  • Loading branch information
tsloughter committed Jun 14, 2024
1 parent 354ffe4 commit 6c39f44
Showing 1 changed file with 24 additions and 10 deletions.
34 changes: 24 additions & 10 deletions projects/convenience.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,19 +2,33 @@

## Background and description

Convenience as a SIG was initially proposed by [Ted Young](https://github.com/tedsuo) back in 2021. It
was postponed to prioritize completion of signal specs and stabilization. A
document was written at this time describing user research to do and a
brainstorming session,
https://docs.google.com/document/d/1bjiw5L4E8FztRtfsy4Nu9A6qSP-BJlK6BBzl6Rea-Ds/edit#heading=h.b7qckyjgu28i
Convenience as a SIG was initially proposed by [Ted
Young](https://github.com/tedsuo) back in 2021. It was postponed to prioritize
completion of signal specs and stabilization. A document was written at this
time describing the intent to do user research and a brainstorming session,
https://docs.google.com/document/d/1bjiw5L4E8FztRtfsy4Nu9A6qSP-BJlK6BBzl6Rea-Ds/edit#heading=h.b7qckyjgu28i.
This brainstorm resulted in a list of some existing convenience functions
languages have and proposed ones, like a `start` which takes the tracer from the
active span in the current context, and proposals for defining annotations for
easier instrumentation.

### Current challenges

Currently the various language implementations resolve user convenience
individually and this can lead to drift between different implementations and
with the spec. An example would be how to handle marking a code block as
`untraced` which has implementations in at least Java, Ruby and Javascript, each
done differently, named differently and with no specification.
We know users can find the API and SDK to be complex and daunting, the challenge
is to collect actionable feedback on this issues. That information is spread
across various languages' issues, blog posts and other online commentary.
Performing an end user survey will help collect some of this knowledge into a
central place that is actionable.

When developer experience issues are encountered and addressed they are
currently done in the various language implementations individually and this can
lead to drift between different implementations and with the spec. An example
would be how to handle marking a code block as `untraced` which has
implementations in at least Java, Ruby and Javascript, each done differently,
named differently and with no specification. This means users working across
languages experience different results for similar features and other language
SIG's aren't necessarily even aware of the pain point that their users may
actually be experiencing.

### Goals, objectives, and requirements

Expand Down

0 comments on commit 6c39f44

Please sign in to comment.