On Mon, Mar 9, 2026 at 10:23 AM Andrey Semashev via Boost < boost@lists.boost.org> wrote:

On March 9, 2026 5:39:24 PM Peter Dimov via Boost <boost@lists.boost.org> wrote: ...

tl;dr for Peter: Low visibility produces few early adopters. Few early adopters produce thin reviews. Thin reviews produce costly libraries. Costly libraries produce low visibility. Four lines in the release notes break the cycle at the cheapest point. wall of text for Andrey: I do hear the scope objection, but I think the premise is narrower than the release notes actually are. The release notes have never been limited to code changes in the tarball. Every release notes page has a "New Libraries" section that reports an ecosystem event - a library passed review during this cycle. The notes also list removed libraries, updated tools, and known issues. The scope already includes "things that happened to Boost since the last release." An endorsed library is one step earlier in the same pipeline. Endorsement is a formal community action with a public record on this list. Reporting it is no different in kind from reporting that a new library was accepted or a tool was updated. Now the practical argument. Peter, you wrote in October 2023: "Reviews from users count more than reviews from non-users, of course." And just last November: "We're between a rock and a hard place here; on one hand, we suffer from a lack of volunteers who would want to manage reviews." If reviews from users carry more weight, and we don't have enough reviewers, then anything that increases the number of people who have actually used a library before its review directly improves the process you're defending. Andrey, you diagnosed this problem sixteen years ago. In March 2010 you wrote: "Boost is rather closed to its community. I don't know how it happens, but on independent news I regularly read of such projects as KDE, GNOME, Qt, Linux Kernel and others, but nearly nothing about Boost. The Boost web site changes rarely - essentially, the news column only lists recent Boost releases. For an outsider, nothing really happens around Boost, and that's sad. If Boost was more open and communicated to the public, I think, there would be much more activity in Boost, and during reviews in particular." That is exactly the problem this proposal addresses. The release notes are the one document where outsiders actually look. A library that arrives at formal review with zero early adopters produces a worse review. A worse review produces a library that costs more to maintain after acceptance. The release notes are the only document with enough readership to materially change the size of that early-adopter pool. Putting this information anywhere else is choosing a smaller audience for the information that feeds review quality. Andrey, you also pointed out in that same thread: "We are also suffering from a lack of reviewers. While we all understand time pressures and the need to complete paying work, the strength of Boost is based on the detailed and informed reviews submitted by you." Four lines in the release notes - name, one-line description, link to the endorsement thread, link to the repository - is a low-cost way to put endorsed libraries in front of exactly those people. The proposed section would be clearly separated from the rest of the notes. It does not editorialize. It reports a fact: these libraries were endorsed by this community since the last release. I think the benefit to review quality justifies four lines in a document whose scope already includes ecosystem events. Vinnie

On March 9, 2026 10:52:26 PM Vinnie Falco <vinnie.falco@gmail.com> wrote:

tl;dr for Peter:

Low visibility produces few early adopters. Few early adopters produce thin reviews. Thin reviews produce costly libraries. Costly libraries produce low visibility.

Four lines in the release notes break the cycle at the cheapest point.

Unrelated content makes release notes less useful to the intended reader, to the point of being annoying.

wall of text for Andrey:

I do hear the scope objection, but I think the premise is narrower than the release notes actually are. The release notes have never been limited to code changes in the tarball. Every release notes page has a "New Libraries" section that reports an ecosystem event - a library passed review during this cycle.

No, it reports what's been added to the release. If a library has passed the review but has not been included in the release (e.g. due to a review ending too late), it wouldn't be mentioned in the notes.

The notes also list removed libraries, updated tools, and known issues. The scope already includes "things that happened to Boost since the last release."

Not what happened but what changed in the new release. Endorsement is not such a change.

An endorsed library is one step earlier in the same pipeline. Endorsement is a formal community action with a public record on this list.

To be frank, I never fully understood the point of endorsement. If I endorse a library, does it mean I'm obligated to write a review? It's a difficult promise to make since endorsements are made before the review time frame is set. And since I bothered to endorse, does it mean I promise my review will be positive? I don't remember examples of the opposite. I don't think the answers to the above questions should be "yes". I guess, you can consider endorsements as an indication of community interest, but not more than that. There are many ways to express interest, why not report on that? If anything, I would rather the repotring increased the interest to the proposed library, regardless of endorsements. A news item on the web site about a proposed library with a short description and an invitation to discuss and review it would be a good thing.

Reporting it is no different in kind from reporting that a new library was accepted or a tool was updated.

Yes, but where that reporting is done matters. Release notes is not the right place, as endorsements are not relevant to the purpose of that document.

Andrey, you diagnosed this problem sixteen years ago. In March 2010 you wrote: "Boost is rather closed to its community. I don't know how it happens, but on independent news I regularly read of such projects as KDE, GNOME, Qt, Linux Kernel and others, but nearly nothing about Boost. The Boost web site changes rarely - essentially, the news column only lists recent Boost releases. For an outsider, nothing really happens around Boost, and that's sad. If Boost was more open and communicated to the public, I think, there would be much more activity in Boost, and during reviews in particular."

That is exactly the problem this proposal addresses. The release notes are the one document where outsiders actually look.

That's not *why* they look in that document. They come there for one purpose and the document must serve that purpose well. You've made a new beautiful website, use it to better expose proposed libraries. You have socials to notify people.

В письме от вторник, 10 марта 2026 г. 14:32:30 MSK пользователь Glen Fernandes via Boost написал:

On Tuesday, March 10, 2026, Andrey Semashev via Boost I haven't seen any library not being endorsed recently, so if the step in the process is largely performative we can decide if it serves any real purpose today.

I remember at least one library proposal in the last few years that wasn't endorsed. So, I think endorsement still has some value.

On March 10, 2026 2:47:28 PM Peter Dimov via Boost <boost@lists.boost.org> wrote:

Andrey Semashev wrote:

...

To be frank, I never fully understood the point of endorsement.

The point of endorsement was to prevent libraries that weren't ready for formal review to have such scheduled and thereby waste everyone's time.

I don't remember there being an influx of underprepared libraries. IIRC, endorsements were an attempt to prevent review periods with no submitted reviews. We didn't have those for a while, but we still did have review periods with little reviews, so I'm not sure how successful that attempt was. Anyway, as I said, there are many ways to express interest in the library and a (non-binding) intention to review it. It's the formality of the act of endorsement and what it entails for the one who endorses is what I don't like or understand.

В письме от вторник, 10 марта 2026 г. 11:56:54 MSK пользователь Andrey Semashev via Boost написал:

Unrelated content makes release notes less useful to the intended reader, to the point of being annoying.

As a sort of middle ground: we can have a link to the News section of the website on release notes page. It could be visually distinct and the text could identify it as something not release-related. Something like "Also check out other Boost news".

On March 10, 2026 8:06:38 PM Dmitry Arkhipov via Boost <boost@lists.boost.org> wrote:

В письме от вторник, 10 марта 2026 г. 11:56:54 MSK пользователь Andrey Semashev via Boost написал:

...

Unrelated content makes release notes less useful to the intended reader, to the point of being annoying.

As a sort of middle ground: we can have a link to the News section of the website on release notes page. It could be visually distinct and the text could identify it as something not release-related. Something like "Also check out other Boost news".

There already are links at the top of the release notes page, including News.

No dia 10 de mar. de 2026, às 21:03, Marcelo Zimbres Silva via Boost <boost@lists.boost.org> escreveu:

On Mon, 9 Mar 2026 at 13:33, Vinnie Falco via Boost <boost@lists.boost.org> wrote:

...

The release notes currently cover changes to libraries shipped in each Boost release. I'd like to propose adding a small, clearly separated section called "Newly Endorsed Libraries" that lists libraries endorsed on this mailing list since the previous release.

I support this. An endorsement is the first item in the list of things a library needs to get into boost, If the results of a review are included it seems natural to also include the endorsement.

Review results are not included in the release notes currently. Joaquín M López Muñoz

On Wed, Mar 11, 2026 at 1:04 AM Ion Gaztañaga <igaztanaga@gmail.com> wrote:

...

I really just want as many users as possible going into the review. I don't care how it happens. Release Notes is an obvious Schelling point. A lot of people read them and they are a natural audience (they are already Boost users). Thanks