April 2023 - Boost - lists.preview.boost.org

Re: [boost] Boost.Compat
by Marshall Clow 27 Apr '23

27 Apr '23

On Apr 27, 2023, at 8:31 AM, Glen Fernandes via Boost <boost(a)lists.boost.org> wrote: > > On Thu, Apr 27, 2023 at 11:24 AM Peter Dimov via Boost < > boost(a)lists.boost.org> wrote: > >> It would be nice if we had a dedicated place in Boost where to >> put these, to avoid the need for everyone reimplementing a >> private copy. >> >> At the moment, some of these have gone into Core, for lack of >> anywhere else, and we now have a <latch> contribution (which I >> encouraged): >> >> https://github.com/boostorg/core/pull/144 >> >> However, Core is not quite the right place for these components. >> >> So my suggestion is to create a separate library/repo, Boost.Compat >> (boostorg/compat, namespace boost::compat) for them. >> >> Thoughts? (Be sure to read the discussion in the PR as well.) >> > > > For the rest of the list: I'm in favor of this (as my response in the > Core thread indicates). There’s a lot of this kind of stuff in Boost.Algorithm Boost/algorithm/cxx11, for example. — Marshall

1 0

Re: [boost] Boost.Compat
by Glen Fernandes 27 Apr '23

27 Apr '23

On Thu, Apr 27, 2023 at 11:24 AM Peter Dimov via Boost < boost(a)lists.boost.org> wrote: > It would be nice if we had a dedicated place in Boost where to > put these, to avoid the need for everyone reimplementing a > private copy. > > At the moment, some of these have gone into Core, for lack of > anywhere else, and we now have a <latch> contribution (which I > encouraged): > > https://github.com/boostorg/core/pull/144 > > However, Core is not quite the right place for these components. > > So my suggestion is to create a separate library/repo, Boost.Compat > (boostorg/compat, namespace boost::compat) for them. > > Thoughts? (Be sure to read the discussion in the PR as well.) > For the rest of the list: I'm in favor of this (as my response in the Core thread indicates). Glen

1 0

Re: [boost] Boost.Compat
by Niall Douglas 27 Apr '23

27 Apr '23

On 27/04/2023 16:16, Peter Dimov via Boost wrote: > However, Core is not quite the right place for these components. > > So my suggestion is to create a separate library/repo, Boost.Compat > (boostorg/compat, namespace boost::compat) for them. I'd prefer if those polyfills were categorised according to C++ standard, so: Boost.CXX11 Boost.CXX14 Boost.CXX17 Boost.CXX20 Boost.CXX23 ... Then when a person upgrades the minimum C++ standard for a codebase, what needs to be switched to std:: is very easy to identify. Niall

1 0

Boost.Compat
by Peter Dimov 27 Apr '23

27 Apr '23

Newer C++ standards (20) have been adding useful things to the standard library (<bit>, <latch>, etc) and most of us would like to use them, both in Boost and outside of it, while still supporting C++11, or 14, or 17. Many of these components are straightforward to implement using the C++11 standard library, consist of not much than one relatively small header, don't need many Boost dependencies beyond the required ones (Assert, ThrowException, Config), and since their interface is already fixed by the standard, don't need a design review. It would be nice if we had a dedicated place in Boost where to put these, to avoid the need for everyone reimplementing a private copy. At the moment, some of these have gone into Core, for lack of anywhere else, and we now have a <latch> contribution (which I encouraged): https://github.com/boostorg/core/pull/144 However, Core is not quite the right place for these components. So my suggestion is to create a separate library/repo, Boost.Compat (boostorg/compat, namespace boost::compat) for them. Thoughts? (Be sure to read the discussion in the PR as well.)

1 0

Re: [boost] Replacing the ML with a forum
by Vinnie Falco 27 Apr '23

27 Apr '23

On Thu, Apr 27, 2023 at 7:56 AM Robert Ramey via Boost <boost(a)lists.boost.org> wrote: > Boost has a history making it's own tools: > ... > It's not a great history. In fact, I spend more time getting boost > tools to work than I do working on boost software. The difference here is that our efforts are powered by staff engineers who draw a salary and external corporations who have, as a condition of payment, the accomplishment of our goals. Thanks

1 0

Re: [boost] Replacing the ML with a forum
by Vinnie Falco 27 Apr '23

27 Apr '23

On Thu, Apr 27, 2023 at 7:56 AM Robert Ramey via Boost <boost(a)lists.boost.org> wrote: > I fail to see what is to be gained by replacing the current system. The current system is not being "replaced" it is being enhanced, with the OPTION (but not the requirement) to interact with the mailing lists using a modern, forum-based interface. Just as there are folks who have expressed that they would not participate in Boost discussions outside of the mailing list interface, there are as many (or more) people who have expressed that they will not interact with Boost through the mailing list at all. You don't hear from them because well, they don't post to the list (the dog that didn't bark). You ask, what is to be gained? Well, the largest existential threat to Boost is that we do not refresh the foundations of our social technology and our membership. Library membership in Boost no longer carries the cachet that it used to. Authors of new libraries increasingly prefer to go directly to the C++ standard instead of being thoroughly vetted in a Boost review. If we want to continue to maintain Boost's extraordinary relevance, we must do the following: 1. Make the mailing list more appealing to younger generations (millenials and the TikTok youths) 2. Increase participation in the Boost Formal Review process 3. Spark the imagination of the new generations of library authors - to write new libraries - to submit their libraries to Boost for review - to become active maintainers upon acceptance 4. Inspire people to contribute to Boost - by using the libraries, and providing feedback - by contributing to individual projects - by helping with documentation - by helping other users with questions or problems The mailing lists must never go away, for as long as there are even a few people who prefer them. But this cannot be the sole method to participate in the Boost social space. To stay current we must provide additional choices. The need for a forum interface is obvious. Thanks

1 0

Re: [boost] Replacing the ML with a forum
by Robert Ramey 27 Apr '23

27 Apr '23

On 4/26/23 8:02 AM, Vinnie Falco via Boost wrote: > - No existing solution meets our needs, so we have to write it ourselves The above sentence is a huge red flag. Boost has a history making it's own tools: a) documentation chain b) build system - Bjam c) build system - CMake integration c) CI d) GitHub integration d) probably some others I forgot It's not a great history. In fact, I spend more time getting boost tools to work than I do working on boost software. Like most computer projects, it turns out to be a lot of work which drags on forever and is constantly fiddled with to make work. When the work becomes tedious enthusiasm peters out and were stuck with something that "can be made to work". I fail to see what is to be gained by replacing the current system. Robert Ramey

1 0

Re: [boost] Replacing the ML with a forum
by Gavin Lambert 26 Apr '23

26 Apr '23

On 27/04/2023 08:40, Robert Ramey wrote: > I use a threaded email reader - Thunderbird. > > I get my mailing list posts via a news reader. > > This is a perfectly wonderful setup and I'd be loath to change it. +1 on all three points. Via gmane, specifically. Thus far a newsreader has been the only successful way I can follow a community -- flat mailing lists are too disjointed and hard to follow context, and websites are "out of sight, out of mind"; I might keep up for a few weeks but inevitably forget for a while and then I'm 1k messages behind and I just ignore it forever.

1 0

proposal Multi multidimensional array library
by Alfredo Correa 26 Apr '23

26 Apr '23

Hi all, My name is Alfredo. I would like to propose this library called Multi for inclusion in Boost. Multi is a container and views library for (dynamic) multidimensional arrays. The library is available here, https://gitlab.com/correaa/boost-multi. The Boost prefix is purely aspirational at this point, of course. As I understand the process, the important step at this point is to find someone(s) to endorse the library and later a review manager. If you want to be an endorser, please let me know. If anyone has feedback or tips to prepare it for a review, please let me know as well. I will be at the Aspen CPPNOW conference if someone wants to discuss it in person. I am in the C++ Slack channel #boost for discussions as well. Here is a link to try it quickly online https://godbolt.org/z/KerbxW6WK. Thank you, Alfredo

1 0

Re: [boost] Replacing the ML with a forum
by Robert Ramey 26 Apr '23

26 Apr '23

On 4/26/23 7:08 AM, Vinnie Falco via Boost wrote: > Before we replace the mailing list with a forum (and implement a mail > gateway) can I get feedback on how you use the mailing list from your > email client? > > Are you using a threaded e-mail reader? If so, do you care if > discussions are changed to be flat (like Discourse)? What is important > to you (besides that you can continue to use your email client)? > I use a threaded email reader - Thunderbird. I get my mailing list posts via a news reader. This is a perfectly wonderful setup and I'd be loath to change it. Robert Ramey

1 0