PageSpeed August 2020 report draft

[Otto van der Schaaf <os...@we-amp.com>]

Hi all,

I have posted the nearly due board report draft over at
https://cwiki.apache.org/confluence/display/INCUBATOR/August2020

Please take a look; feedback is highly appreciated

Otto

Re: PageSpeed August 2020 report draft

[Nick Kew <ni...@apache.org>]

> On 4 Aug 2020, at 22:16, Otto van der Schaaf <os...@we-amp.com> wrote:
> 
> Thank you Nick, I followed up on the draft and incorporated some of your
> feedback.
> For the WIP-part of DISCLAIMER, I think what's left are trivial things to
> fix - except for the handling
> of the crypto part of the code base.

Cheers.

Actually it occurs to me, that wasn't so much feedback on the report itself,
rather it was comments on the points you raise.

> We depend on openssl/boringssl for some things, and I'm just not familiar
> with what should be done here / what it takes to obtain an export license
> for that.

IANAL, but surely you're not exporting them, you're just linking to them?
They're already part of all the underlying servers, with or without pagespeed.
We have long-established precedent for that @apache!

-- 
Nick Kew

Re: PageSpeed August 2020 report draft

[Otto van der Schaaf <os...@we-amp.com>]

Thank you Nick, I followed up on the draft and incorporated some of your
feedback.
For the WIP-part of DISCLAIMER, I think what's left are trivial things to
fix - except for the handling
of the crypto part of the code base.

We depend on openssl/boringssl for some things, and I'm just not familiar
with what should be done here / what it takes to obtain an export license
for that.

There might also be another road towards compliance: to axe out features
that induce the ssl dependency
(and maybe allow a plugin model to wire back in some of that if someone
desires so, e.g. for https fetching
of input resources)


On Mon, Aug 3, 2020 at 1:05 PM Nick Kew <ni...@apache.org> wrote:

>
>
> > On 3 Aug 2020, at 11:30, Otto van der Schaaf <os...@we-amp.com> wrote:
> >
> > Hi all,
> >
> > I have posted the nearly due board report draft over at
> > https://cwiki.apache.org/confluence/display/INCUBATOR/August2020
> >
> > Please take a look; feedback is highly appreciated
> >
> > Otto
>
> Feedback:
>
> It's quite terse throughout.  Which is probably fine, except where the
> jargon
> goes over my head ("... stand-alone sidecar service"?)
>
> The interesting part is all in one dense paragraph:
>
> "The first incubator release has been announced after field-testing.
> Furthermore, small enhancements are being contributed from people outside
> of the initial committer group. Major changes have landed on master: the
> build system has been refreshed to leverage bazel, c++17, absl & Envoy. As
> a side-effect, most of the remaining issues from the WIP-DISCLAIMER have
> been addressed."
>
> External contributors are good: are they potential internal contributors?
> That is to say,
> consider whether anyone should be committers: build the community.  Though
> of course
> discussion of individuals takes place in private@ until&unless they are
> invited and accept.
>
> If WIP-Disclaimer can be dropped in the near future, that's something to
> raise here:
> summarise what the issues were and how they've been addressed.  It's also
> a box
> to tick on the way to graduation.
>
> The later parts of the report appear to be TBD.
>
> --
> Nick Kew

Re: PageSpeed August 2020 report draft

[Nick Kew <ni...@apache.org>]

> On 3 Aug 2020, at 11:30, Otto van der Schaaf <os...@we-amp.com> wrote:
> 
> Hi all,
> 
> I have posted the nearly due board report draft over at
> https://cwiki.apache.org/confluence/display/INCUBATOR/August2020
> 
> Please take a look; feedback is highly appreciated
> 
> Otto

Feedback:

It's quite terse throughout.  Which is probably fine, except where the jargon
goes over my head ("... stand-alone sidecar service"?)

The interesting part is all in one dense paragraph:

"The first incubator release has been announced after field-testing. Furthermore, small enhancements are being contributed from people outside of the initial committer group. Major changes have landed on master: the build system has been refreshed to leverage bazel, c++17, absl & Envoy. As a side-effect, most of the remaining issues from the WIP-DISCLAIMER have been addressed."

External contributors are good: are they potential internal contributors?  That is to say,
consider whether anyone should be committers: build the community.  Though of course
discussion of individuals takes place in private@ until&unless they are invited and accept.

If WIP-Disclaimer can be dropped in the near future, that's something to raise here:
summarise what the issues were and how they've been addressed.  It's also a box
to tick on the way to graduation.

The later parts of the report appear to be TBD.

-- 
Nick Kew