Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager. Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do? This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse? Discuss. -- Regards, Vinnie Follow me on GitHub: https://github.com/vinniefalco
I’m more than happy to rewrite parts of the Boost website with, perhaps, bootstrap and mustaches. It would be fitting, I think, to use boost-based projects for the website’s back-end I’m currently working on a boost::beast-based server. It hosts my website, but that’s not exactly mass-tested. I am curious in re. the thought that Boost is on a decline. If the point of comparison is high-impact library creation and adoption, then yes, maybe boost is on a decline. But, aren’t there only so many fundamental libraries to create — and doesn’t boost cover virtually all of those fundamental libraries? There might be room to pivot in the direction of more user-facing utilities. Things like graphics clients, window managers, or audio I/O come to mind. There may also be room to pivot in the opposite direction: embedded utilities. Things such as GPIO, SPI, and I2C come to mind as easy candidates for such a library. The state of that landscape is a bit scattered right now. On one hand, the Linux kernel gives the programmer robust support for such connections — but because all of the support is file-based, programmers generally pick a library with less overhead. That library used to be wiringpi, which has since gone dark. Developers may find such projects attractive. If my experience is any indication of how other programmers think, I’ll put it out there: the existing boost libraries are either too well-written (beast), too large, or too domain-specific (math libraries) for me to contribute anything of substance. WL
On Apr 4, 2022, at 1:43 PM, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do?
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
Discuss.
-- Regards, Vinnie
Follow me on GitHub: https://github.com/vinniefalco
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On Mon, Apr 4, 2022 at 12:05 PM William Linkmeyer <wlink10@gmail.com> wrote:
It would be fitting, I think, to use boost-based projects for the website’s back-end I’m currently working on a boost::beast-based server. It hosts my website, but that’s not exactly mass-tested.
While I am flattered, trying to use Boost.Beast and C++ to build a modern website may not be the most efficient. That's a lot of new code which few people will understand. I haven't deployed any servers recently, but I suspect that the economical and pragmatic solution is to use off-the-shelf software parts that are widely understood and for which it is easy to find reasonably skilled maintainers.
I am curious in re. the thought that Boost is on a decline. If the point of comparison is high-impact library creation and adoption, then yes, maybe boost is on a decline. But, aren’t there only so many fundamental libraries to create — and doesn’t boost cover virtually all of those fundamental libraries?
Human creativity is essentially infinite, and software is infinitely frustrating, so I think that we will not be covering all the potentially useful libraries any time soon, if ever :) You do raise a good question, how do we measure the success of Boost? Is it the number of installs? The number of programs using Boost? The amount of participation on the lists? I think software development is an inherently social phenomenon. It is the vibrant exchange of knowledge and ideas that drives the state-of-the-art forward. Thanks
On Mon, 4 Apr 2022 at 21:11, William Linkmeyer via Boost < boost@lists.boost.org> wrote:
I’m more than happy to rewrite parts of the Boost website with, perhaps, bootstrap and mustaches.
It would be fitting, I think, to use boost-based projects for the website’s back-end I’m currently working on a boost::beast-based server. It hosts my website, but that’s not exactly mass-tested.
I am curious in re. the thought that Boost is on a decline. If the point of comparison is high-impact library creation and adoption, then yes, maybe boost is on a decline. But, aren’t there only so many fundamental libraries to create — and doesn’t boost cover virtually all of those fundamental libraries?
There might be room to pivot in the direction of more user-facing utilities. Things like graphics clients, window managers, or audio I/O come to mind.
I favour this. Boost is great for supplying nuts and bolts, but it's light on turnkey application frameworks.
There may also be room to pivot in the opposite direction: embedded utilities. Things such as GPIO, SPI, and I2C come to mind as easy candidates for such a library. The state of that landscape is a bit scattered right now. On one hand, the Linux kernel gives the programmer robust support for such connections — but because all of the support is file-based, programmers generally pick a library with less overhead. That library used to be wiringpi, which has since gone dark.
Developers may find such projects attractive. If my experience is any indication of how other programmers think, I’ll put it out there: the existing boost libraries are either too well-written (beast), too large, or too domain-specific (math libraries) for me to contribute anything of substance.
WL
On Apr 4, 2022, at 1:43 PM, Vinnie Falco via Boost < boost@lists.boost.org> wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do?
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
Discuss.
-- Regards, Vinnie
Follow me on GitHub: https://github.com/vinniefalco
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
If boost were able to take this concept: https://ossia.io/posts/minimum-viable/ And drive it further, we may be able to place roots in graphics for later expansion. I anticipate further community involvement after the above idea is refined. It’s a great launch-point into a system-agnostic [ui/wm/graphics] framework. Google’s flutter and filament come to mind as the kinds of projects that would be possible for Boost after this — a minimum viable declarative graphics library (which I’ll call “DGL” for brevity) — is in place. The potential applications are quite broad. Building on Beast, DGL could also be served to clients on the web and declare user interfaces users there. WL
On Apr 4, 2022, at 4:28 PM, Richard Hodges via Boost <boost@lists.boost.org> wrote:
On Mon, 4 Apr 2022 at 21:11, William Linkmeyer via Boost < boost@lists.boost.org> wrote:
I’m more than happy to rewrite parts of the Boost website with, perhaps, bootstrap and mustaches.
It would be fitting, I think, to use boost-based projects for the website’s back-end I’m currently working on a boost::beast-based server. It hosts my website, but that’s not exactly mass-tested.
I am curious in re. the thought that Boost is on a decline. If the point of comparison is high-impact library creation and adoption, then yes, maybe boost is on a decline. But, aren’t there only so many fundamental libraries to create — and doesn’t boost cover virtually all of those fundamental libraries?
There might be room to pivot in the direction of more user-facing utilities. Things like graphics clients, window managers, or audio I/O come to mind.
I favour this. Boost is great for supplying nuts and bolts, but it's light on turnkey application frameworks.
There may also be room to pivot in the opposite direction: embedded utilities. Things such as GPIO, SPI, and I2C come to mind as easy candidates for such a library. The state of that landscape is a bit scattered right now. On one hand, the Linux kernel gives the programmer robust support for such connections — but because all of the support is file-based, programmers generally pick a library with less overhead. That library used to be wiringpi, which has since gone dark.
Developers may find such projects attractive. If my experience is any indication of how other programmers think, I’ll put it out there: the existing boost libraries are either too well-written (beast), too large, or too domain-specific (math libraries) for me to contribute anything of substance.
WL
On Apr 4, 2022, at 1:43 PM, Vinnie Falco via Boost < boost@lists.boost.org> wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do?
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
Discuss.
-- Regards, Vinnie
Follow me on GitHub: https://github.com/vinniefalco
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On 2022-04-04 15:05, William Linkmeyer via Boost wrote:
It would be fitting, I think, to use boost-based projects for the website’s back-end
Don't forget to also distribute boost-based web browsers for users to use when visiting the website. You could add some checks into the HTTP headers being exchanged. Or in fact, perhaps it's time to consider a new set of boost-based protocols to replace HTTP, while we are considering an update ? Heck, isn't this entire discussion four days late ? Stefan -- ...ich hab' noch einen Koffer in Berlin...
On 04/04/2022 18:43, Vinnie Falco via Boost wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
The last time there was a period of declining activity both on the mailing list and the lack of new libraries I decided to do something about it, and based on results achieved I'd say it got turned around. However unless there is somebody actively stoking the input fires, the pipeline doesn't pump itself, and things stop moving along.
Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do?
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
Similar ideas were floated last time round. What brings life to Boost is people seeing it as relevant and current. If they think it is relevant and current, they participate. If they don't, they stay silent. I think that's the case no matter the website nor the discussion forum. Note how Outcome's review drew hundreds of responses on this mailing list. As did Beast's. As did Hana's. Those libraries engaged the wider community. Get reviews pumping, then people will participate. Without changes worth commenting upon, people won't engage. To get reviews pumping, you need a combination of new libraries which are attractive enough, and willing review managers. I solved that last time by drawing in young engineers who had the fresh ideas to make interesting libraries, and the free time to get them past review. I also spent a lot of time on the conference circuit cajoling people to act as review managers. I stopped attending conferences four years ago, around the same time I stopped being Boost's Google Summer of Code admin. Niall
On 2022-04-04 13:43, Vinnie Falco via Boost wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do?
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
It stands to reason that a decline of participation (or development activity in general) is somehow caused by the website not being attractive or the communication requiring the wrong tools. In fact, I remember numerous discussions all being misguided by the underlying assumption that everything can be solved by new and shiny tools. (What else would you expect from a bunch of engineers !?) (For avoidance of doubt, I have nothing against a website overhaul or a move to a different communication channel. But I strongly doubt that this per se will have any lasting effect on the community.) Stefan -- ...ich hab' noch einen Koffer in Berlin...
On 4/4/22 20:43, Vinnie Falco via Boost wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
We've had this discussion before and I stand by my opinion. I find forums vastly inferior to mailing lists for discussions.
I have no answers, but maybe some things to highlight: * We've written a lot of stuff, it's mostly well written, sometimes even maintained, but the point is we haven't left many obvious "holes" to fill, especially holes that could be filled by newcomers. To pick a random example, type-traits used to generate a steady flow of "what about this trait" or "what about this compiler workaround" messages. In the great scheme of things these were not earth shattering requests, but they were actually great for getting people engaged, and allowing new participants the opportunity to cut their teeth on a tricky problem. Better compilers (no more SFINAE workarounds - yeh!!) and a pretty comprehensive set of traits means those kinds of messages just don't show up anymore. This is great for library maintenance, but rubbish for group-cohesion. * "Balkanization" of Boost: For most of the stuff I maintain, messages go straight to Github as either PR's or Issues, and discussion takes place there. Now don't get me wrong... I actually like Github's workflow, but it does mean that unless you're interested in say Math or Multiprecision and choose to follow the issues there, the discussion all takes place in our own little group. * In relation to the above - *everyone* has something to contribute to *every* library - let me give you an example - some of Math's contributors are keen to explore C++20 modules, the question is, how on earth do we go about it and maintain a single code base, plus first class module support and backwards compatible headers? It's something that needs a Boost-wide discussion and maybe some guidance and experimentation. Maybe we need a list of open questions - some would relate to new compiler features - concepts and modules spring to mind, some would relate to missing libraries. BTW I'm neutral on the whole message-board vs mailing list issue.... to me the medium isn't necessarily the problem. John.
Viewpoint from a user of Boost. I'm firmly in the camp that a forum is the way to go. Boost has been crying out for this for AGES! Pete On Mon, 4 Apr 2022 at 18:44, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
Clearly, a website update is necessary and along with it some type of campaign to highlight once again the important role that Boost serves in the C++ community for both professionals and amateurs alike. But what more can we do?
This came up before but it is worth mentioning again; in addition to a website update to make boost.org modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list? I realize this will ruffle some feathers but surely the alternative, a descent into irrelevance due to steady declining activity is worse?
Discuss.
-- Regards, Vinnie
Follow me on GitHub: https://github.com/vinniefalco
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On 5/04/2022 20:31, Peter Barker via Boost wrote:
Viewpoint from a user of Boost. I'm firmly in the camp that a forum is the way to go. Boost has been crying out for this for AGES!
I'm in the opposite camp. Invariably any community that has switched from newsgroups/mailing lists to a web forum has caused me to forget that it even exists in fairly short order. Not deliberately, just... out of sight, out of mind, y'know? I'm still using gmane for this list, in fact, because it does threading better than plain email seems to manage. And "better threading" seems to be one of the common factors for a request to move to forums. So perhaps the Real Solution™ is to move back to newsgroups! (This was mostly a joke. Mostly.)
Vinnie Falco wrote:
Clearly, a website update is necessary
Why, exactly? I mean, how many people visit the website but then decide not to use Boost because they don't like the font?
how do we feel about a transition to forum-based discussion instead of the mailing list?
I think the question we need to ask first is "where have the participants gone?". Preferably quantified. One place I've noticed is that people go to the std:: mailing lists with their ideas and bypass Boost.
Follow me on GitHub: https://github.com/vinniefalco
Yeah, that's another place some people have gone. Compared to just publishing your project on github, what does Boost add? Github has the advantage that you don't need to endure a review and maybe be forced to change things! And, as John Maddock has pointed out, Boost has already adopted github and some of what used to happen on this list now happens there. Robert Ramey wrote:
you might want to bring this up in a meeting at C++Now
That's another place that some people have gone. Personally I think it's important for Boost not to be "two-tier" i.e. those with travel budgets and those without. Regards, Phil.
On 4/5/22 14:39, Phil Endecott via Boost wrote:
I think the question we need to ask first is "where have the participants gone?". Preferably quantified.
Another place is reddit. There are reddit posts about Boost that also ought to have been posted here. Off the top of my head: https://www.reddit.com/r/cpp/comments/t8wlqs/development_plan_for_boostunord... https://www.reddit.com/r/cpp/comments/k58a2u/beman_dawes_has_passed_away_boo... https://www.reddit.com/r/cpp/comments/giu6xg/proposal_for_an_epochbased_orga...
On Tue, 5 Apr 2022 at 16:45, Bjorn Reese via Boost <boost@lists.boost.org> wrote:
On 4/5/22 14:39, Phil Endecott via Boost wrote:
I think the question we need to ask first is "where have the participants gone?". Preferably quantified.
Another place is reddit. There are reddit posts about Boost that also ought to have been posted here.
This is a good point - it is now the norm for social media management to be an important part of any web presence.
Off the top of my head:
https://www.reddit.com/r/cpp/comments/t8wlqs/development_plan_for_boostunord...
https://www.reddit.com/r/cpp/comments/k58a2u/beman_dawes_has_passed_away_boo...
https://www.reddit.com/r/cpp/comments/giu6xg/proposal_for_an_epochbased_orga...
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On Mon, 4 Apr 2022 at 19:43, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
Clearly, a website update is necessary [...]
I agree. Given the number of users I encountered lost*** on spaces (Gitter, Slack) of Geometry or GIL libraries asking for general Boost or library-specific directions, that I have helped lately, tells me t is not too easy (esp. for newcomers) to figure where ask for help and report problems too. *** Two days ago, a user interested in Boost.Graph was asking questions on Gitter room of Boost.Geometry. Now, if it is easier to find that Gitter room than Boost mailing list or Graph's issues, then something is wrong indeed about the Boost frontdoor signs.
how do we feel about a transition to forum-based discussion instead of the mailing list?
I like mailing lists, but I understand NKOTB-s may prefer the web. How about the GitHub Discussions? How many Boost libraries have enabled them? Has it attracted the discussions? Best regards, -- Mateusz Loskot, http://mateusz.loskot.net
On 04/04/2022 18:43, Vinnie Falco via Boost wrote:
how do we feel about a transition to forum-based discussion instead of the mailing list?
How about creating a free newsgroups on alt.* or free.* ( I don't recommend comp.* because they are very bureaucratic and very difficult to create new groups on them). Newsgroups are easy access using any news client such as Thunderbird or something else on Linux. Newsgroups are free and are carried by many news servers free of charge. The only snag could be trolls posting their spam posts but if anything is to go by then there are C and C++ newsgroups where trolls and spammers are almost negligible. -- With over 1.7 billion devices now running Windows 10/11, customer satisfaction is higher than any previous version of windows.
On 14/04/2022 07:59, Vinnie Falco wrote:
On Wed, Apr 13, 2022 at 12:27 PM 😉 Good Guy 😉 wrote:
How about creating a free newsgroups on alt.* or free.*
Newsgroups are still around?
I still read this list via the gmane newsgroup :P If you want to do that too, it's gmane.comp.lib.boost.devel on news.gmane.io.
This feels like one step forward, two steps sideways. I’m going to put a +1 out for Fossil-scm and a +1 out for GitHub discussions. I’m in favor of simplifying the pipelines. Make it *obvious* where things are. WL
On Apr 13, 2022, at 7:33 PM, Gavin Lambert via Boost <boost@lists.boost.org> wrote:
On 14/04/2022 07:59, Vinnie Falco wrote:
On Wed, Apr 13, 2022 at 12:27 PM 😉 Good Guy 😉 wrote: How about creating a free newsgroups on alt.* or free.* Newsgroups are still around?
I still read this list via the gmane newsgroup :P
If you want to do that too, it's gmane.comp.lib.boost.devel on news.gmane.io.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On 14/04/2022 11:50, William Linkmeyer wrote:
This feels like one step forward, two steps sideways.
I’m going to put a +1 out for Fossil-scm and a +1 out for GitHub discussions.
I’m in favor of simplifying the pipelines. Make it *obvious* where things are.
Well, to me it's obvious: things are here. Either as an email list or as a newsgroup; each user can decide for themselves and they're fully interchangeable. I've never used Slack or Reddit or whatever else. GitHub comments are fine, but they should be focused on a specific issue or PR, not for more general discussion.
On 4/13/22 4:32 PM, Gavin Lambert via Boost wrote:
On 14/04/2022 07:59, Vinnie Falco wrote:
On Wed, Apr 13, 2022 at 12:27 PM 😉 Good Guy 😉 wrote:
How about creating a free newsgroups on alt.* or free.*
Newsgroups are still around?
I still read this list via the gmane newsgroup :P
If you want to do that too, it's gmane.comp.lib.boost.devel on news.gmane.io.
+1 for news groups. Slack is a poor substitute.
On Mon, Apr 4, 2022 at 7:43 PM Vinnie Falco via Boost <boost@lists.boost.org> wrote:
how do we feel about a transition to forum-based discussion instead of the mailing list?
Yes please. I don't really participate a lot here, but one of the worst things is inability to have nicely formatted code, maybe embedded godbolt preview etc... Mailing list really does feel* 20 years ago tech*. If there is some little fundraiser to make this happen I would be happy to donate. I know some people may be used to mailing lists, but I really do hope this move forward ;)happens. As for why I am not contributing: not good enough :), and lack of time, but those are things on my side so lets focus on what Boost could improve: I think the most annoying thing is that there is no clear flow for me to do small improvements since IDK what libs are dead and which are not, or what are the good issues for casual contributors, e.g. something that most regular experienced C++ devs can do, so you do not need to be pdimov or ldionne to do it... E.g. I want to add initializer list to boost::bimap. Ups, somebody did that already, ups their work is wasted because lib is unmaintained. 😥 https://github.com/boostorg/bimap/pull/31/files
On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list?
Strongly opposed. The beauty of email is that * the discussion *comes to me*; that I don’t have to go and check some website to see if there are new posts. * I have a local archive of the boost mailing lists going back to (checks…) 2009 at my fingertips. Pre-2009 requires a bit more work. * I can read (and search, and reply) with tools of my choice. If you know of a forum-based discussion that offers those benefits, I’d love to hear about it. — Marshall
On Tue, Apr 5, 2022, 6:41 PM Marshall Clow via Boost <boost@lists.boost.org> wrote:
On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> modern and
relevant, how do we feel
about a transition to forum-based discussion instead of the mailing list?
Strongly opposed.
The beauty of email is that
* the discussion *comes to me*; that I don’t have to go and check some website to see if there are new posts. * I have a local archive of the boost mailing lists going back to (checks…) 2009 at my fingertips. Pre-2009 requires a bit more work. * I can read (and search, and reply) with tools of my choice.
If you know of a forum-based discussion that offers those benefits, I’d love to hear about it.
On Apr 5, 2022, at 5:07 PM, René Ferdinand Rivera Morell <grafikrobot@gmail.com> wrote:
On Tue, Apr 5, 2022, 6:41 PM Marshall Clow via Boost <boost@lists.boost.org <mailto:boost@lists.boost.org>> wrote: On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org <mailto:boost@lists.boost.org>> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> <http://boost.org/ <http://boost.org/>> modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list?
Strongly opposed.
The beauty of email is that
* the discussion *comes to me*; that I don’t have to go and check some website to see if there are new posts. * I have a local archive of the boost mailing lists going back to (checks…) 2009 at my fingertips. Pre-2009 requires a bit more work. * I can read (and search, and reply) with tools of my choice.
If you know of a forum-based discussion that offers those benefits, I’d love to hear about it.
I forgot the most important thing about email. * It doesn’t interrupt me when “new stuff is available”; I can look at it when I want. I looked at Discourse a year ago, and hated it. I didn’t see any information about alternate clients, workflows etc a year ago, so I’ll take a look again, but the my take on the default experience was “like Slack, but with cheesier graphics and more sounds, animations, and interruptions”. — Marshall
On Tue, Apr 5, 2022 at 7:55 PM Marshall Clow <mclow.lists@gmail.com> wrote:
On Apr 5, 2022, at 5:07 PM, René Ferdinand Rivera Morell < grafikrobot@gmail.com> wrote:
I forgot the most important thing about email.
* It doesn’t interrupt me when “new stuff is available”; I can look at it when I want.
I looked at Discourse a year ago, and hated it.
I didn’t see any information about alternate clients, workflows etc a year ago, so I’ll take a look again, but the my take on the default experience was “like Slack, but with cheesier graphics and more sounds, animations, and interruptions”.
It supports email list integration. I.e. it supports all the same features you listed because you can use it through an email list. From the features list:
Reply via email When you aren’t active on the website, your notifications will be automatically sent to you via email. Reply via email from anywhere, on any device.
Mailing list support
I'm not going to claim I like it. Just saying software that meets your criteria exists ;-) -- -- René Ferdinand Rivera Morell -- Don't Assume Anything -- No Supone Nada -- Robot Dreams - http://robot-dreams.net
On 4/6/22 03:54, Marshall Clow via Boost wrote:
On Apr 5, 2022, at 5:07 PM, René Ferdinand Rivera Morell <grafikrobot@gmail.com> wrote:
On Tue, Apr 5, 2022, 6:41 PM Marshall Clow via Boost <boost@lists.boost.org <mailto:boost@lists.boost.org>> wrote: On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org <mailto:boost@lists.boost.org>> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> <http://boost.org/ <http://boost.org/>> modern and relevant, how do we feel about a transition to forum-based discussion instead of the mailing list?
Strongly opposed.
The beauty of email is that
* the discussion *comes to me*; that I don’t have to go and check some website to see if there are new posts. * I have a local archive of the boost mailing lists going back to (checks…) 2009 at my fingertips. Pre-2009 requires a bit more work. * I can read (and search, and reply) with tools of my choice.
If you know of a forum-based discussion that offers those benefits, I’d love to hear about it.
I forgot the most important thing about email.
* It doesn’t interrupt me when “new stuff is available”; I can look at it when I want.
I looked at Discourse a year ago, and hated it.
I didn’t see any information about alternate clients, workflows etc a year ago, so I’ll take a look again, but the my take on the default experience was “like Slack, but with cheesier graphics and more sounds, animations, and interruptions”.
I'd like to add another point of consideration. Any centralized service may enact access restrictions based on user's region or other criteria. This is not specifically about Discourse as I don't know if they are practicing this. I know Slack does. Also, web-based forums are more susceptible to user's government access restrictions. Spam, discussions on sensitive or illegal in certain jurisdictions topics and malicious comments stay available online, unless actively moderated, and may be grounds for penalties and access restrictions. Gladly, so far I haven't heard of restrictions applied to email exchange.
On 06/04/2022 01:54, Marshall Clow via Boost wrote:
* It doesn’t interrupt me when “new stuff is available”; I can look at it when I want.
I looked at Discourse a year ago, and hated it.
I didn’t see any information about alternate clients, workflows etc a year ago, so I’ll take a look again, but the my take on the default experience was “like Slack, but with cheesier graphics and more sounds, animations, and interruptions”.
Discourse is absolutely awful. I say this as someone who has been using it for most of the past decade. It requires Javascript enabled to load any content. Search engines - even Google - appear completely incapable of indexing most of its content. It locks the content into database formats which require maintaining. It's a wide open security hole. I think the display of conversations and topics has crap UX. It tries to be IM-y rather that discussion-y, and it ends up being crap at both. In short, I hate Discourse. It has all the overhead and maintenance hassles of phpBB, but with worse UX, worse usability, worse searchability, worse scalability. It's inferior on most metrics. Something tolerable - currently before they ruin it - is Reddit. I particularly like that search engines index it well, and given the useless rubbish Google nowadays returns for a query, I often wish for a Google which exclusively searches Reddit and returns no other results. Its mobile integration is also tolerable, though I uninstalled its app because it spys on you and then its UX tries to refuse to work if it detects a mobile phone browser, because in the end you're not their customer and as usual they're ultimately out to get you. Reddit has no maintenance nor security overheads for us, its main issue is lock in and undoubtedly its owners will at some point ruin it and then all our content will be gone forever. I'm still struggling to find much better than email mailing lists personally. I wish it weren't so, but there's a real gap in the market for decent, modern, non-lock-in, low maintenance, discussion software. Niall
On Wed, Apr 6, 2022 at 10:44 AM Niall Douglas via Boost <boost@lists.boost.org> wrote: > On 06/04/2022 01:54, Marshall Clow via Boost wrote: > >> https://www.discourse.org/ <https://www.discourse.org/> > > * It doesn’t interrupt me when “new stuff is available”; I can look at it when I want. > > I looked at Discourse a year ago, and hated it. > Discourse is absolutely awful. I'd quickly mention (again?) the [Fossil-SCM-based forum][1] used for both SQLite and Fossil itself. Dr Richard Hipp, of SQLite fame, was fed up with fighting spam on the ML, and decided to take matters into his own capable hands, as he often does, and created his own forum software, on top of his own SCM software (Fossil), itself on top of SQLite, itself built on Lemon, Richard's own Yacc replacement. (he almost implemented his own SMTP, and started it in fact, but for now uses the system one AFAIK). Oh, and the [Web server is also based on his own of course][3]. As a long time SQLite ML member, I was strongly against it, especially the anonymous posting, but turns out to be a non-issue, since I can still read messages in plain readable text in my mail client, while replying must be done in the low-key minimal and fast web-app; I actually even enjoy the ability to format things in markup (markdown or fossil varieties), and even have diagrams described in PIC format (i.e. plain text, see [2]). As usual for My Hipp's productions, it's high quality, small, OSS, in plain C (and TH, a TCL subset, for templating), very fast and low resources (so cheap to host). And contrary to SQLite, external contributions are possible, with a larger set of committers, beside Richard who remains the main author and shepherd for the project. These are not commercial projects, Richard has nothing to sell you (beside support contract on SQLite if you want). These tools are first and foremost for Richard himself and SQLite, but happen to be generally useful and mature IMHO. I've used Discord and Basecamp, and don't hate them, but Fossil forums are so lightweight, fast, and low-key, that it's basically the antithesis of these JS-heavy Flashy "modern" forum software, and I prefer Fossil now. And there's no Fossil Forum "app" for phones or tablets. But the site aims to be "responsive" enough to be useable. And again, you also need to use the web-app for posting, not reading. Anyways, I thought I'd be remise not to mention all the above. Even though I doubt Boost would adopt a Fossil forum. FWIW, --DD [1]: https://fossil-scm.org/home/doc/trunk/www/forum.wiki [2]: https://fossil-scm.org/home/doc/trunk/www/pikchr.md [3]: https://sqlite.org/althttpd/doc/trunk/althttpd.md
On 06/04/2022 12:42, Dominique Devienne wrote: > On Wed, Apr 6, 2022 at 10:44 AM Niall Douglas via Boost > <boost@lists.boost.org> wrote: >> On 06/04/2022 01:54, Marshall Clow via Boost wrote: >>>> https://www.discourse.org/ <https://www.discourse.org/> >>> * It doesn’t interrupt me when “new stuff is available”; I can look at it when I want. >>> I looked at Discourse a year ago, and hated it. >> Discourse is absolutely awful. > > I'd quickly mention (again?) the [Fossil-SCM-based forum][1] used for > both SQLite and Fossil itself. It's been a while since I last looked at Fossil, and I like what they've improved. Here is an example of a longer discussion: https://fossil-scm.org/forum/forumpost?name=1dd382219c - Default view is threaded-chronological, but they put buttons on the top to quickly switch between views. - One can quickly toggle Markdown rendering off and on from the top. - Doesn't need Javascript to render. - Looks and works fine on a phone. - Docs claims it works offline from browser cache and will resync next internet connection, handy if you're on a train or a plane. Nice. My biggest issue with everything Dr. Hipp has done outside SQLite is he's fallen into the trap of "one ring to solve all problems" i.e. Fossil implements a SCM, a wiki, issue tracking, forum software, and lots of other stuff, all in one program binary. I don't want lots of stuff in one overarching solution. I want one thing solved well, and I want it to plug nicely into the most popular tooling. Which is git for SCM and github for issue tracking. So that's a big turnoff for me at least, it reminds me of that all-in-one integrated thingy which Boost had yonks ago and we were never able to get off some very ancient version of it full of known security holes. I **really** don't want to go back to that. Niall
On Wed, Apr 6, 2022 at 2:15 PM Niall Douglas via Boost <boost@lists.boost.org> wrote:
On 06/04/2022 12:42, Dominique Devienne wrote:
I'd quickly mention [Fossil-SCM-based forum]
I don't want lots of stuff in one overarching solution. I want one thing solved well, and I want it to plug nicely into the most popular tooling. Which is git for SCM and github for issue tracking.
Well, when that overarching solution weighs less than most Boost libraries :) Some people are using Fossil solely for the forum. Ignore the rest. Heck, you could even subset the code needed, if you wanted. The ML is not integrated with Git or GitHub, yet works fine. Think of it that way.
So that's a big turnoff for me at least, it reminds me of that all-in-one integrated thingy which Boost had yonks ago and we were never able to get off some very ancient version of it full of known security holes. I **really** don't want to go back to that.
It's actually because Mr Hipp cares a lot about security that he controls the whole stack... I can guarantee you, from having witnessed it, that he's very reactive to any report about issues. And if you actually look into the sources of althttpd, you'd see clearly stated the code is kept simple on purpose, to make it auditable from A to Z in a single sitting of a couple hours. All requests are handled in a forked child on purpose (works best/fastest on Linux for that reason), with the whole thing running in a chroot jail by default. I.e. the attack surface is kept small on purpose. I'm sure it could further be secured in a container or nanovm or whatever the Cloud'y people will invent next. Again, I doubt Fossil will gain traction here. And I'm just a (happy) user of it, I have no skin in this game. But please don't go about writing about security holes or abondonware or bloat for things Dr Hipp does...
On 06/04/2022 13:58, Dominique Devienne wrote:
So that's a big turnoff for me at least, it reminds me of that all-in-one integrated thingy which Boost had yonks ago and we were never able to get off some very ancient version of it full of known security holes. I **really** don't want to go back to that.
But please don't go about writing about security holes or abondonware or bloat for things Dr Hipp does...
The issue with the all in one integrated thingy which Boost had yonks ago wasn't that there weren't new versions. Because there were, at least two major releases and countless minor releases. We were stuck on something like v0.5. The issue was that upgrading the existing to newer broke stuff, and nobody was willing to volunteer the time to fix the existing stuff. So we stayed stuck forever on the ancient version. This has absolutely nothing to do with Dr. Hipp. It has everything to do with us locking ourselves into technologies which we then find later we aren't willing to pay the maintenance for, and then we get hoist by our own petard. As much as GitHub is a catastrophic single point of failure, and that used to be unacceptable here as a dependency, once Microsoft took them over we have mostly stopped worrying and learned to love GitHub. I for one like that none of us here have to maintain GitHub, keep it safe from security attacks, and keep it maintained. That's worth a lot of money to us here to not have to do. Niall
On Wed, Apr 6, 2022 at 4:13 PM Niall Douglas via Boost <boost@lists.boost.org> wrote:
On 06/04/2022 13:58, Dominique Devienne wrote:
So that's a big turnoff for me at least, it reminds me of that all-in-one integrated thingy which Boost had yonks ago and we were never able to get off some very ancient version of it full of known security holes. I **really** don't want to go back to that. But please don't go about writing about security holes or abondonware or bloat for things Dr Hipp does...
The issue [...] which Boost had [...] wasn't that there weren't new versions. [...] The issue was that upgrading the existing to newer broke stuff [...]
Good thing Fossil and SQLite have a excellent track record of backward compatibility then.
[...] learned to love GitHub. [...]
But that's orthogonal, no? GitHub does not have MLs or a Boost-wide forum, does it? I don't have experience with GitHub discussions, but I do subscribe to lots of GitHub projects and some issues, getting email notifications from them, but this doesn't feel like a forum or an ML to me. But perhaps I'm missing your point and what you are actually advocating for, as a replacement for the current MLs? In the GitHub ecosystem perhaps? --DD PS: Another anecdote regarding "security" and Fossil. When SHA1 was "broken", Dr Hipp switched Fossil to SHA3 in a matter of days, retaining BC with former artifacts using SHA1, while it took months (or is it years?) for Git to upgrade its hashes. But that's about Fossil *SCM*, while here we are merely discussing Fossil for its Forum feature (yet, Forum posts are SCM artifacts in and of themselves I believe, although I'm not sure of that. They are editable and versioned for sure).
On 06.04.22 14:15, Niall Douglas via Boost wrote:
I don't want lots of stuff in one overarching solution. I want one thing solved well, and I want it to plug nicely into the most popular tooling. Which is git for SCM and github for issue tracking.
Objection! I don't think it's possible for a SCM system to both do SCM well and be compatible with git. That's not doing one thing well, that's trying to do two contradictory things. Fossil does SCM well. I've never used it as a discussion forum or as an issue tracker, so I can't comment on how well it does these things. However, at least neither of these things is in direct contradiction with doing SCM well, or with each other. -- Rainer Deyke (rainerd@eldwood.com)
On 6/04/2022 20:44, Niall Douglas wrote:
Something tolerable - currently before they ruin it - is Reddit. I particularly like that search engines index it well, and given the useless rubbish Google nowadays returns for a query, I often wish for a Google which exclusively searches Reddit and returns no other results.
Just add 'site:reddit.com' to your search queries. Although I think that can still return off-site ads, depending on your search terms.
I would like to throw out some thoughts as one of the newer repeat contributors. 1) The barrier to entry is really high. John Maddock still points out macros or small pieces of legacy wizardry to me, and it's been over 2 years since my first contribution to Boost.Math. There is also the issue of unmaintained libraries. I submitted some trivial changes to Boost.Random that would fix breakage in Math, and it took month's until John overrode the supposed maintainer and merged them in. I think that something to the affect of a Boost version of Kernel Newbies would pay dividends in onboarding new contributors. John, Chris Kormanyos, and Nick Thompson have all been hugely helpful in getting me up to speed but I know it took a lot of time and effort on their end. 2) As others have mentioned I think there is a lot of good conversation that exists in other areas that the mailing list. The mailing list isn't particularly high volume (I have this, the module discussion, and Math/Multiprecision internal as the only 3 active) so I don't know if moving to a forum helps. There also generally seems to be no clear consensus when things get posted to the Mailing List that impact everyone. The one that comes to mind is the discussion of moving on from C++03. I am 29 which means C++11 was finalized when I was a freshman in college. You are not going to attract new people by making them use ancient standards. I appreciate that Math and Multiprecision made the internal decision to move a minimum of C++11 with additional features using 17 (e.g. parallelization). 3) Vinnie posed the questions how do we get more people involved as review managers? I will ask in return how does one get involved as a review manager? I may not be an expert on his proposed URL library, but I certainly have experience in project management. The webpage describing the process seemed a bit nebulous to me, and I am likely not alone in thinking this. 4) Another thought on deepening involvement for people like me is there a list of unmaintained libraries that need someone? What is the process for becoming a maintainer of those or in general? I am pretty sure the only reason John gave me write access to Math was so that I would stop pestering him. What help do other libraries need and what is their glide path? I have spent the better part of the last year and a half working on standalone modes for Math and Multiprecision. The standalone math is now used in the MSVC STL and Multiprecision was done by request. We have all seen the Reddit posts about people not using Boost because it is a monolith so we know there is a large demand signal. If there are other libraries that want to pursue this I am certainly willing to help. I also think an information campaign on moving beyond the monolith would be worthwhile. Again just my 2 cents. I do not intend to undermine the work that has been going on since I was a child. Matt
On Wed, 6 Apr 2022 at 03:29, Matt Borland via Boost <boost@lists.boost.org> wrote:
3) Vinnie posed the questions how do we get more people involved as review managers? I will ask in return how does one get involved as a review manager?
As per https://www.boost.org/community/reviews.html#Review_Manager First, "an active boost member not connected with the library submission must volunteer to be the "Review Manager" for the library." Then, "The library author has to accept a person as a review manager." Finally, the Wizards "Approve the review manager based on initial acceptance by the library submitter, their participation in the Boost community, including the mailing list, previous reviews, and other forums."
The webpage describing the process seemed a bit nebulous to me, and I am likely not alone in thinking this.
The description is pretty complete - I'd risk statement it's one of the best documented Boost-wide processes out there :) Still, I agree, there is room for improvements, e.g. it could explain the process in a cooking recipe-like or algorithmic fashion :) Best regards, -- Mateusz Loskot, http://mateusz.loskot.net
On Wed, 6 Apr 2022 at 02:08, René Ferdinand Rivera Morell via Boost < boost@lists.boost.org> wrote:
Please, do not joke about it :-( We "decided" to switch from an old mailing list software to Discourse in house, and after few weeks of use it seems clear that Discourse is making all the wrong choices regarding "mailing list mode": - it alters the body of the message adding "fancy" images (with tracking links) - it puts the recipient as the sole address in the "to" field, making it harder to filter emails based on the recipient list - it uses a different reply-to email address for each "thread", making it hard to filter emails based on that, too - it does not provide simple email addresses to start new threads - it does not keep the sender in cc or any other header, preventing one from writing to the sender If you were to seriously considering it, the LLVM people seem to be intent in switching to it, and I would suggest to wait and see how that plays out :-/ Best, .Andrea
On Wed, Apr 6, 2022 at 2:39 AM Andrea Bocci <andrea.bocci@cern.ch> wrote:
On Wed, 6 Apr 2022 at 02:08, René Ferdinand Rivera Morell via Boost < boost@lists.boost.org> wrote:
Please, do not joke about it :-(
We "decided" to switch from an old mailing list software to Discourse in house, and after few weeks of use it seems clear that Discourse is making all the wrong choices regarding "mailing list mode": - it alters the body of the message adding "fancy" images (with tracking links) - it puts the recipient as the sole address in the "to" field, making it harder to filter emails based on the recipient list - it uses a different reply-to email address for each "thread", making it hard to filter emails based on that, too - it does not provide simple email addresses to start new threads - it does not keep the sender in cc or any other header, preventing one from writing to the sender
If you were to seriously considering it, the LLVM people seem to be intent in switching to it, and I would suggest to wait and see how that plays out :-/
Thank you for that. Others have suggested Discourse, like for WG21, but I've never gotten to experience it. I wasn't a fan of the web UX style at first glance. And combined with your assessment cements my opinion against it :-) -- -- René Ferdinand Rivera Morell -- Don't Assume Anything -- No Supone Nada -- Robot Dreams - http://robot-dreams.net
On Tue, Apr 5, 2022 at 4:41 PM Marshall Clow via Boost < boost@lists.boost.org> wrote:
On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> modern and
relevant, how do we feel
about a transition to forum-based discussion instead of the mailing list?
Strongly opposed.
Also strongly opposed. I am now the maintainer of Boost for Fedora/CentOS/RHEL (Jonathan is transition out of the role). If Boost discussion moves to some forum that I have to chase down vs. just arriving alongside the rest of my upstreams (e.g. libstdc++) it's likely to get missed, so the quality of Fedora's distribution of Boost will likely suffer. The beauty of email is that
* the discussion *comes to me*; that I don’t have to go and check some website to see if there are new posts. * I have a local archive of the boost mailing lists going back to (checks…) 2009 at my fingertips. Pre-2009 requires a bit more work. * I can read (and search, and reply) with tools of my choice.
If you know of a forum-based discussion that offers those benefits, I’d love to hear about it.
— Marshall
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On 06/04/2022 17:48, Thomas Rodgers via Boost wrote:
On Tue, Apr 5, 2022 at 4:41 PM Marshall Clow via Boost < boost@lists.boost.org> wrote:
On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> modern and
relevant, how do we feel
about a transition to forum-based discussion instead of the mailing list?
Strongly opposed.
Also strongly opposed. I am now the maintainer of Boost for Fedora/CentOS/RHEL (Jonathan is transition out of the role). If Boost discussion moves to some forum that I have to chase down vs. just arriving alongside the rest of my upstreams (e.g. libstdc++) it's likely to get missed, so the quality of Fedora's distribution of Boost will likely suffer.
As kind of proof of the problem, a lot of the discussion around this topic has been occurring on Slack on #boost, rather than here. To sum up the discussion there, the more instant discussion happens on Slack, increasingly less immediate discussion is also happening on github discussions, and of course github issues. In other words, discussion around Boost increasingly: - Is centred per boostorg repo on github (library specific). - Occurs on Slack (instant) or on github boostorg repo (less instant). - Non-per-library discussion tends to happen increasingly on Reddit /r/cpp. Which is to say, discussion is not happening here on this mailing list, or on other mailing lists. So TBH Thomas the ship is sailing here already. The question is whether to actually do anything about it, or leave trends continue uninterrupted i.e. we're clearly already moving onto Github, Reddit, and Slack. Niall
On Wed, Apr 6, 2022 at 10:25 AM Niall Douglas via Boost < boost@lists.boost.org> wrote:
On 06/04/2022 17:48, Thomas Rodgers via Boost wrote:
On Tue, Apr 5, 2022 at 4:41 PM Marshall Clow via Boost < boost@lists.boost.org> wrote:
On Apr 4, 2022, at 10:43 AM, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
This came up before but it is worth mentioning again; in addition to a website update to make boost.org <http://boost.org/> modern and
relevant, how do we feel
about a transition to forum-based discussion instead of the mailing list?
Strongly opposed.
Also strongly opposed. I am now the maintainer of Boost for Fedora/CentOS/RHEL (Jonathan is transition out of the role). If Boost discussion moves to some forum that I have to chase down vs. just arriving alongside the rest of my upstreams (e.g. libstdc++) it's likely to get missed, so the quality of Fedora's distribution of Boost will likely suffer.
As kind of proof of the problem, a lot of the discussion around this topic has been occurring on Slack on #boost, rather than here.
To sum up the discussion there, the more instant discussion happens on Slack, increasingly less immediate discussion is also happening on github discussions, and of course github issues.
In other words, discussion around Boost increasingly:
- Is centred per boostorg repo on github (library specific).
- Occurs on Slack (instant) or on github boostorg repo (less instant).
- Non-per-library discussion tends to happen increasingly on Reddit /r/cpp.
Which is to say, discussion is not happening here on this mailing list, or on other mailing lists.
So TBH Thomas the ship is sailing here already. The question is whether to actually do anything about it, or leave trends continue uninterrupted i.e. we're clearly already moving onto Github, Reddit, and Slack.
Well, my willingness to chase ships is limited...GitHub is the least bad alternative for my purposes, IMO.
Niall
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On Wed, Apr 6, 2022 at 10:31 AM Thomas Rodgers via Boost <boost@lists.boost.org> wrote:
Well, my willingness to chase ships is limited...GitHub is the least bad alternative for my purposes, IMO.
I agree that you should not have to "chase ships." If Boost moves to a discussion forum I think it only makes sense that mailing list functionality is preserved, even if only for a handful of users. Especially if those users have an outsized impact on Boost (such as being a packager or distribution maintainer). Thanks
On Wed, Apr 6, 2022 at 9:58 AM Thomas Rodgers via Boost <boost@lists.boost.org> wrote:
If Boost discussion moves to some forum that I have to chase down vs. just arriving alongside the rest of my upstreams (e.g. libstdc++) it's likely to get missed, so the quality of Fedora's distribution of Boost will likely suffer.
Yep, I think this is more evidence that mailing-list based participation is beneficial when the recipient is involved in multiple communities. A forum solution has to provide a way to opt-in to email participation (subscribe to one or more sub-forums, with the ability to reply) for the subset of users that want consolidated social activity. Thanks
On Apr 4, 2022, at 20:43, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
One way forward might be involving more women. Now and again, a woman posts a patch or a comment but the core participants are all men. The Foundation's board is also all male. Google Code also seems to attract an all-male crowd. Women, in particular, are excellent at active networking and will in time involve yet other women (and men). How to do it? Next time there is a review, cold-call a colleague or a mailing list participant. Cheers, Kostas ========================================= Institute of Nuclear and Particle Physics NCSR Demokritos http://inspirehep.net/author/profile/K.G.Savvidy.1 https://mixmax.hepforge.org
On 4/7/22 09:54, Kostas Savvidis via Boost wrote:
On Apr 4, 2022, at 20:43, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
One way forward might be involving more women. Now and again, a woman posts a patch or a comment but the core participants are all men. The Foundation's board is also all male. Google Code also seems to attract an all-male crowd.
Women, in particular, are excellent at active networking and will in time involve yet other women (and men).
How to do it? Next time there is a review, cold-call a colleague or a mailing list participant.
Gender has no bearing in Boost community, and I would very much like it to stay this way. If you're good at C++, if you have interesting ideas, if you make useful libraries - you are welcome, regardless of gender, race or nationality.
On 07/04/2022 09:37, Andrey Semashev via Boost wrote:
On 4/7/22 09:54, Kostas Savvidis via Boost wrote:
On Apr 4, 2022, at 20:43, Vinnie Falco via Boost <boost@lists.boost.org>
wrote:
the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
One way forward might be involving more women. Now and again, a woman posts a patch or a comment but the core participants are all men. The Foundation's board is also all male. Google Code also seems to attract an all-male crowd.
Women, in particular, are excellent at active networking and will in time involve yet other women (and men).
How to do it? Next time there is a review, cold-call a colleague or a mailing list participant.
Gender has no bearing in Boost community, and I would very much like it to stay this way. If you're good at C++, if you have interesting ideas, if you make useful libraries - you are welcome, regardless of gender, race or nationality.
There seems to be about a 5-10% stable population proportion of women over time, very similar proportion to Physics. Many years ago - maybe ten now? - I did an informal survey at a conference of what puts women off a career in C++. What I say next is as reliable as that would mean, plus with a decade of memory corruption. It's not the technical complexity - actually, that was rated usually as a positive. It's not toxic masculinity - a few mentioned the C++ male dominated ecosystem is marginally better than wider society. It's not outright discrimination, it was mentioned a few misoygnists exist anywhere you go, and C++ was neither better nor worse in that regard than elsewhere. However there was a perception of there being a glass ceiling in terms of promotion into technical specialisation, and much less so into managerial specialisation. Many commented that they were sure they didn't get pay rises anything like as much as men, and that was offputting. From what I gathered at the time, the overwhelming feedback was that it the lack of flexibility shown by employers that put women off. C++ employers want people for at least 40 hours per week. They are usually flexible about when, but not about how much. I heard several horror stories of employer mistreatment and intransigence, admittedly they were all US based, I can't imagine any of that being legal in the EU. I'm sympathetic to that observation. Yesterday morning I had a sick one year old to look after while my wife slept. You don't get much high quality C++ written when there is a one year old actively trying to choke themselves, pull things down onto themselves, and trying to push the power button on your computer. I wasn't able to say "my child is sick, there is nobody else to look after them, so I need to take half a day at very short notice". Even in the EU, there is no legal obligation on employers to provide that right, and the EU is probably the most worker friendly large region anywhere in the world. Re: Foundation/Conferences/WG21 I can absolutely assure you that every effort is made to ensure as diverse a board membership as possible, but there are very few women to choose from, and when asked to serve many say they can't commit the time (which is the case for most approached whether male or female). So if a board ends up all male, it's because there wasn't much alternative. I know for a fact that that lack of diversity bothers everybody involved. In any case, I agree with Andrey, I don't want to know anything about a C++ contributor other than the C++ they have to show me. I also think that if employers were actually serious about improving diversity, they could easily do so, but continue to proactively choose to keep the current situation the way it is. We in open source have very little power over the economics here because we don't have the power to hire and fire, and we have no control over the money. We are takers here, not setters. All that said, god speed on improving the situation. Niall
On Thu, Apr 7, 2022 at 4:37 AM Andrey Semashev wrote:
On 4/7/22 09:54, Kostas Savvidis via Boost wrote:
One way forward might be involving more women. Now and again, a woman posts a patch or a comment but the core participants are all men. The Foundation's board is also all male. Google Code also seems to attract an all-male crowd.
Women, in particular, are excellent at active networking and will in time involve yet other women (and men).
Gender has no bearing in Boost community, and I would very much like it to stay this way. If you're good at C++, if you have interesting ideas, if you make useful libraries - you are welcome, regardless of gender, race or nationality.
+1. Glen
On Thu, 7 Apr 2022 at 10:37, Andrey Semashev via Boost <boost@lists.boost.org> wrote:
On 4/7/22 09:54, Kostas Savvidis via Boost wrote:
On Apr 4, 2022, at 20:43, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
One way forward might be involving more women. Now and again, a woman posts a patch or a comment but the core participants are all men. The Foundation's board is also all male. Google Code also seems to attract an all-male crowd.
Women, in particular, are excellent at active networking and will in time involve yet other women (and men).
How to do it? Next time there is a review, cold-call a colleague or a mailing list participant.
Gender has no bearing in Boost community, and I would very much like it to stay this way. If you're good at C++, if you have interesting ideas, if you make useful libraries - you are welcome, regardless of gender, race or nationality.
+1 I couldn't express my own thoughts on that matter better. Best regards, -- Mateusz Loskot, http://mateusz.loskot.net
On Thu, Apr 7, 2022 at 1:41 PM Mateusz Loskot via Boost <boost@lists.boost.org> wrote:
On Thu, 7 Apr 2022 at 10:37, Andrey Semashev via Boost
Gender has no bearing in Boost community
+1 I couldn't express my own thoughts on that matter better.
If gender was such a non-issue, then we wouldn't have [affirmative action][1] (in general) or [gender-quotas-by-law][2] in our societies. Doing nothing, or saying there's no issue, is basically been "totally fine" with the "status quo". That's OT and I have no clue how such things apply to volunteer-based organizations like Boost, but lets be honest, the status quo rarely changes w/o being "challenged" a little, often my law. --DD [1] https://conseilsdejournalistes.com/en/egalite-genre/13-quest-ce-que-la-discr... [2] https://www.forbes.com/sites/avivahwittenbergcox/2021/05/15/france-unanimous...
On Thu, 7 Apr 2022 at 14:16, Dominique Devienne <ddevienne@gmail.com> wrote:
On Thu, Apr 7, 2022 at 1:41 PM Mateusz Loskot via Boost <boost@lists.boost.org> wrote:
On Thu, 7 Apr 2022 at 10:37, Andrey Semashev via Boost
Gender has no bearing in Boost community
+1 I couldn't express my own thoughts on that matter better.
If gender was such a non-issue, then we wouldn't have [affirmative action][1] (in general) or [gender-quotas-by-law][2] in our societies.
Perhaps other organisations experienced issues, so they take actions as they see it appropriate. I do not need to be presented with rules of ethics written down, - no one needs it in here, I'm certain - in order to behave ethically and professionally. Best regards, -- Mateusz Loskot, http://mateusz.loskot.net
On 4/7/22 15:19, Dominique Devienne via Boost wrote:
On Thu, Apr 7, 2022 at 1:41 PM Mateusz Loskot via Boost <boost@lists.boost.org> wrote:
On Thu, 7 Apr 2022 at 10:37, Andrey Semashev via Boost
Gender has no bearing in Boost community
+1 I couldn't express my own thoughts on that matter better.
If gender was such a non-issue, then we wouldn't have [affirmative action][1] (in general) or [gender-quotas-by-law][2] in our societies.
My personal opinion is that any quota-based approach is detrimental to the community. Not just Boost community, but in general. The moment you start segregating people by qualities irrelevant to that community focus and goals, you are introducing social divide in the community and you are taking away from its efficiency.
Hi, On Thu, 7 Apr 2022 at 14:16, Dominique Devienne via Boost <boost@lists.boost.org> wrote:
On Thu, Apr 7, 2022 at 1:41 PM Mateusz Loskot via Boost <boost@lists.boost.org> wrote:
If gender was such a non-issue, then we wouldn't have [affirmative action][1] (in general) or [gender-quotas-by-law][2] in our societies. Doing nothing, or saying there's no issue, is basically been "totally fine" with the "status quo". That's OT and I have no clue how such things apply to volunteer-based organizations like Boost, but lets be honest, the status quo rarely changes w/o being "challenged" a little, often my law. --DD
[1] https://conseilsdejournalistes.com/en/egalite-genre/13-quest-ce-que-la-discr... [2] https://www.forbes.com/sites/avivahwittenbergcox/2021/05/15/france-unanimous...
I am very skeptical any of this applies to online discussions. How can an online male be intimidating to a woman? I suppose physical presence would be a prerequisite for that. I don't even know the gender of some names I see popping up here. Marcelo
On Thu, Apr 7, 2022 at 1:41 PM Mateusz Loskot via Boost <boost@lists.boost.org> wrote:
On Thu, 7 Apr 2022 at 10:37, Andrey Semashev via Boost
Gender has no bearing in Boost community +1 I couldn't express my own thoughts on that matter better. If gender was such a non-issue, then we wouldn't have [affirmative action][1] (in general) or [gender-quotas-by-law][2] in our societies. Doing nothing, or saying there's no issue, is basically been "totally fine" with the "status quo". That's OT and I have no clue how such things apply to volunteer-based organizations like Boost, but lets be honest, the status quo rarely changes w/o being "challenged" a little, often my law. --DD
[1] https://conseilsdejournalistes.com/en/egalite-genre/13-quest-ce-que-la-discr... [2] https://www.forbes.com/sites/avivahwittenbergcox/2021/05/15/france-unanimous... The fact that some organization or a government wants to follow some value system (religion or ideology) doesn't mean that others have to do
W dniu 07.04.2022 o 14:19, Dominique Devienne via Boost pisze: this as well and that the outcome would be positive in the long term. What is being asked here is to abandon merit as the most important value and instead to focus on equality of outcome based on immutable characteristics instead. So I propose to stop with the nice words about "challenging" and talk about what's really happening here. Regards, Adam P.S. The parliament (unanimously) mandating something is the status quo. Politics is downstream from culture, not the other way around.
Respectfully, this seems to be getting slightly off-topic. I enjoyed the discussions about the benefits of Fossil-scm. Although, I think (or hope) that we can all agree that there could be efforts to make most any open-source community be more welcoming. Two cents, WL
On Apr 7, 2022, at 9:18 AM, Adam Wulkiewicz via Boost <boost@lists.boost.org> wrote:
W dniu 07.04.2022 o 14:19, Dominique Devienne via Boost pisze:
On Thu, Apr 7, 2022 at 1:41 PM Mateusz Loskot via Boost <boost@lists.boost.org> wrote: On Thu, 7 Apr 2022 at 10:37, Andrey Semashev via Boost
Gender has no bearing in Boost community +1 I couldn't express my own thoughts on that matter better. If gender was such a non-issue, then we wouldn't have [affirmative action][1] (in general) or [gender-quotas-by-law][2] in our societies. Doing nothing, or saying there's no issue, is basically been "totally fine" with the "status quo". That's OT and I have no clue how such things apply to volunteer-based organizations like Boost, but lets be honest, the status quo rarely changes w/o being "challenged" a little, often my law. --DD
[1] https://conseilsdejournalistes.com/en/egalite-genre/13-quest-ce-que-la-discr... [2] https://www.forbes.com/sites/avivahwittenbergcox/2021/05/15/france-unanimous... The fact that some organization or a government wants to follow some value system (religion or ideology) doesn't mean that others have to do this as well and that the outcome would be positive in the long term. What is being asked here is to abandon merit as the most important value and instead to focus on equality of outcome based on immutable characteristics instead. So I propose to stop with the nice words about "challenging" and talk about what's really happening here.
Regards, Adam
P.S. The parliament (unanimously) mandating something is the status quo. Politics is downstream from culture, not the other way around.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On Thu, 7 Apr 2022 at 15:29, William Linkmeyer via Boost <boost@lists.boost.org> wrote:
Although, I think (or hope) that we can all agree that there could be efforts to make most any open-source community be more welcoming.
Yes, let's identify and eliminate technical barriers and chores that work against growing participation by users and contributors. Best regards, -- Mateusz Loskot, http://mateusz.loskot.net
W dniu 07.04.2022 o 15:32, Mateusz Loskot via Boost pisze:
Although, I think (or hope) that we can all agree that there could be efforts to make most any open-source community be more welcoming. Yes, let's identify and eliminate technical barriers and chores
On Thu, 7 Apr 2022 at 15:29, William Linkmeyer via Boost <boost@lists.boost.org> wrote: that work against growing participation by users and contributors. +1
Adam
Hi, I've attempted to synthesize this discussion so we might find holes in our divergencies or come up with something more actionable. - This list does not represent my opinions. I marked things mentioned as pros and cons with +1/-1. This is also how I interpreted the comments and not my opinion. - As with any synthesis, there is going to be some error. I'm sorry if I misrepresented some opinions by mistake. - Solutions are not exclusive and not exhaustive. ## Communication Problems mentioned: - Declining activity in the mailing list - Balkanization of Boost Proposed solutions: - Mailing list - +1 The medium is not necessarily the problem - -1 Declining activity already - -1 Discussions about libraries are already on Github - -1 Instant discussions are already on Slack - -1 _Some_ non-per-library discussions are already on Reddit - -1 Feels link 20-years ago tech - +1 Discussion comes to the subscribers - +1 Simplifies local archives and one can use tools of choice - -1 Inability to have nicely formatted code - -1 Inability to have embedded godbolt preview - +1 It doesn’t interrupt you when “new stuff is available” - Forum based solution - +1 Friendlier from the user perspective - +1 Allows one to subscribe to topics of interest - +1 Discussion also comes to the subscribers - -1 Some find forums inferior to mailing lists - -1 Not everything can be solved by new and shiny tools - -1 People have already moved discussions to Github - +1 Newcomers prefer the web - -1 People need to subscribe to topics - -1 Integration with email involves lots of important details - -1 Centralized services might limit access based on the user's region - -1 Forums might require javascript to load content - Some form of social media management Related points - Getting reviews pumping should incentivize communication ## Promotion Problems mentioned: - Necessity to highlight once again the important role that Boost serves in the C++ community - Newcomers can't figure out where to ask for help and report problems. - Newcomers find it easier to find support on Reddit, Gitter, Github, or slack, than on the mailing list. Proposed solutions: - Website - +1 Make boost.org more modern and relevant - -1 Not everything can be solved by new and shiny tools - -1 Not many people decide not to use Boost because of the website - +1 Could simplify involvement for newcomers - Use boost projects for website backend - -1 May not be the most efficient - -1 Off-the-shelf software parts are widely understood - Some type of campaign - Identify metrics of success - Number of installs - Number of programs using Boost - Participation - Quantify where the participants have gone - People go to std:: mailing lists with their ideas and bypass boost - People go to Github, have their ideas available, bypass the review, and avoid changing things. ## Contributions and Proposals Problems mentioned: - No high-impact library creation and adoption - High barrier to getting involved - Libraries are either too well-written (beast) to contribute - Libraries are too domain-specific (math libraries) to contribute - There is no clear flow to do small improvements - Contributors don't know what libs are dead and which are not - Contributors don't know what are good issues for casual participants - Unmaintained libraries and PRs that get ignored - Contributors' lack of time - You are not going to attract new people by making them use ancient standards - Not enough women Proposed solutions: - Pivot in the direction of more user-facing utilities - +1 Graphics clients, window managers, audio I/O - +1 Boost is great for supplying nuts and bolts, but it's light on turnkey application frameworks - Pivot in the direction of more embedded utilities - GPIO, SPI, and I2C - Drawing in young engineers with fresh ideas and free time - Somebody making that happen - Create a list of open questions - Spend time at conferences cajoling people to act as review managers - +1 Good opportunity to find people - -1 Boost should not be "two-tier" i.e. those with travel budgets and those without - Involve more women: Cold-call a colleague or a mailing list participant - -1 Gender has no bearing in the Boost community - -1 Information about a C++ contributor other than their code is irrelevant - +1 Ignoring this as an issue is being "fine" with the "status quo" - -1 It segregates people by qualities irrelevant to the community - -1 It introduces a social divide - -1 It takes away from the community's efficiency - -1 The problems mentioned might not apply to online discussions - -1 It abandons merit as the most important value - -1 There's no reason to follow other organization's value system - +1 There could be efforts to make any open-source community more welcoming ## Review Problems mentioned: - Declining level of participation as reviewers - Difficulty to find a review manager - Newcomers don't understand the review process, although others do - Newcomers don't understand the process to become a maintainer Proposed solutions: - Somebody making that happen Related problems: - Requires new attractive proposals On Thu, 7 Apr 2022 at 10:36, Adam Wulkiewicz via Boost < boost@lists.boost.org> wrote:
W dniu 07.04.2022 o 15:32, Mateusz Loskot via Boost pisze:
Although, I think (or hope) that we can all agree that there could be efforts to make most any open-source community be more welcoming. Yes, let's identify and eliminate technical barriers and chores
On Thu, 7 Apr 2022 at 15:29, William Linkmeyer via Boost <boost@lists.boost.org> wrote: that work against growing participation by users and contributors. +1
Adam
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
-- Alan Freitas https://github.com/alandefreitas
Hello, I’d like to resurface this thread. There’s a lot of potential here. Alan’s synthesis is clear and actionable. It seems to me that the least controversial improvement we can make is this:
- Pivot in the direction of more user-facing utilities
- +1 Graphics clients, window managers, audio I/O
- +1 Boost is great for supplying nuts and bolts, but it's light on turnkey application frameworks
WL
On Apr 7, 2022, at 2:40 PM, Alan de Freitas via Boost <boost@lists.boost.org> wrote:
Hi,
I've attempted to synthesize this discussion so we might find holes in our divergencies or come up with something more actionable.
- This list does not represent my opinions. I marked things mentioned as pros and cons with +1/-1. This is also how I interpreted the comments and not my opinion. - As with any synthesis, there is going to be some error. I'm sorry if I misrepresented some opinions by mistake. - Solutions are not exclusive and not exhaustive.
## Communication
Problems mentioned:
- Declining activity in the mailing list
- Balkanization of Boost
Proposed solutions:
- Mailing list
- +1 The medium is not necessarily the problem
- -1 Declining activity already
- -1 Discussions about libraries are already on Github
- -1 Instant discussions are already on Slack
- -1 _Some_ non-per-library discussions are already on Reddit
- -1 Feels link 20-years ago tech
- +1 Discussion comes to the subscribers
- +1 Simplifies local archives and one can use tools of choice
- -1 Inability to have nicely formatted code
- -1 Inability to have embedded godbolt preview
- +1 It doesn’t interrupt you when “new stuff is available”
- Forum based solution
- +1 Friendlier from the user perspective
- +1 Allows one to subscribe to topics of interest
- +1 Discussion also comes to the subscribers
- -1 Some find forums inferior to mailing lists
- -1 Not everything can be solved by new and shiny tools
- -1 People have already moved discussions to Github
- +1 Newcomers prefer the web
- -1 People need to subscribe to topics
- -1 Integration with email involves lots of important details
- -1 Centralized services might limit access based on the user's region
- -1 Forums might require javascript to load content
- Some form of social media management
Related points
- Getting reviews pumping should incentivize communication
## Promotion
Problems mentioned:
- Necessity to highlight once again the important role that Boost serves in the C++ community
- Newcomers can't figure out where to ask for help and report problems.
- Newcomers find it easier to find support on Reddit, Gitter, Github, or slack, than on the mailing list.
Proposed solutions:
- Website
- +1 Make boost.org more modern and relevant
- -1 Not everything can be solved by new and shiny tools
- -1 Not many people decide not to use Boost because of the website
- +1 Could simplify involvement for newcomers
- Use boost projects for website backend
- -1 May not be the most efficient
- -1 Off-the-shelf software parts are widely understood
- Some type of campaign
- Identify metrics of success
- Number of installs
- Number of programs using Boost
- Participation
- Quantify where the participants have gone
- People go to std:: mailing lists with their ideas and bypass boost
- People go to Github, have their ideas available, bypass the review, and avoid changing things.
## Contributions and Proposals
Problems mentioned:
- No high-impact library creation and adoption
- High barrier to getting involved
- Libraries are either too well-written (beast) to contribute
- Libraries are too domain-specific (math libraries) to contribute
- There is no clear flow to do small improvements
- Contributors don't know what libs are dead and which are not
- Contributors don't know what are good issues for casual participants
- Unmaintained libraries and PRs that get ignored
- Contributors' lack of time
- You are not going to attract new people by making them use ancient standards
- Not enough women
Proposed solutions:
- Pivot in the direction of more user-facing utilities
- +1 Graphics clients, window managers, audio I/O
- +1 Boost is great for supplying nuts and bolts, but it's light on turnkey application frameworks
- Pivot in the direction of more embedded utilities
- GPIO, SPI, and I2C
- Drawing in young engineers with fresh ideas and free time
- Somebody making that happen
- Create a list of open questions
- Spend time at conferences cajoling people to act as review managers
- +1 Good opportunity to find people
- -1 Boost should not be "two-tier" i.e. those with travel budgets and those without
- Involve more women: Cold-call a colleague or a mailing list participant
- -1 Gender has no bearing in the Boost community
- -1 Information about a C++ contributor other than their code is irrelevant
- +1 Ignoring this as an issue is being "fine" with the "status quo"
- -1 It segregates people by qualities irrelevant to the community
- -1 It introduces a social divide
- -1 It takes away from the community's efficiency
- -1 The problems mentioned might not apply to online discussions
- -1 It abandons merit as the most important value
- -1 There's no reason to follow other organization's value system
- +1 There could be efforts to make any open-source community more welcoming
## Review
Problems mentioned:
- Declining level of participation as reviewers
- Difficulty to find a review manager
- Newcomers don't understand the review process, although others do
- Newcomers don't understand the process to become a maintainer
Proposed solutions:
- Somebody making that happen
Related problems:
- Requires new attractive proposals
On Thu, 7 Apr 2022 at 10:36, Adam Wulkiewicz via Boost < boost@lists.boost.org> wrote:
W dniu 07.04.2022 o 15:32, Mateusz Loskot via Boost pisze:
Although, I think (or hope) that we can all agree that there could be efforts to make most any open-source community be more welcoming. Yes, let's identify and eliminate technical barriers and chores
On Thu, 7 Apr 2022 at 15:29, William Linkmeyer via Boost <boost@lists.boost.org> wrote: that work against growing participation by users and contributors. +1
Adam
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
-- Alan Freitas https://github.com/alandefreitas
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On Thu, 7 Apr 2022 at 20:40, Alan de Freitas via Boost <boost@lists.boost.org> wrote:
- Pivot in the direction of more user-facing utilities - +1 Graphics clients, window managers, audio I/O - +1 Boost is great for supplying nuts and bolts, but it's light on turnkey application frameworks
Generally, I agree. I hope recent proposals of the MySQL and Redis libraries make up the steps in that direction. Best regards, -- Mateusz Loskot, http://mateusz.loskot.net p.s. Who remembers std::rdb from the BoostCon'09 workshop ;)
On Mon, 4 Apr 2022 at 19:43, Vinnie Falco via Boost <boost@lists.boost.org> wrote:
Well, now that I have everyone's attention just before the release, I'd like to draw attention to the elephant in the room. That is, the declining level of activity on the Boost mailing lists and the declining level of participation in the Boost formal review process. Both in terms of the number of reviewers, and in terms of the difficulty in finding a review manager.
Hi, I think there still some points worth being discussed in this thread 1. Libraries that are proposed for boost may not have been written with that intention in mind. That may mean they don't follow the boost format. Boostfying may be too costly since after rejection authors would have to revert back. I think it should be explicitly stated that libraries can't be rejected on the basis of their noncompliance with e.g. namespace boost, boost documentation format and boost build system. 2. Make the review more transparent. If I understand correctly, almost anyone can be a review manager (criteria unclear) and that the acceptance is not decided counting yes and no votes. What happens however when the Review manager is not knowledgeable about the subject? How should he decide whether to accept or not? Authors want a fair treatment of their review. 3. I believe most proposals will come from people that are not paid to develop boost libraries and do it as a side project. That means the review schedule should make it easier for them to engage in the review process. At the moment however I see a 10 days review. Isn't this too short? In my case I would only be able to engage seriously on weekends. 4. I would like to have an easy to find link to a list of libraries that are being written for boost. Perhaps even in the front page. It gives the good impression that the community is thriving. Marcelo
On Tue, Apr 19, 2022 at 1:27 AM Marcelo Zimbres Silva via Boost <boost@lists.boost.org> wrote:
...
These are all great ideas, and it sounds like you are volunteering to help us organize information on the new website :) Please keep a note of these ideas and open them as issues when we deploy a private beta. Thanks
On 4/19/22 01:26, Marcelo Zimbres Silva via Boost wrote: <snip>
Hi,
I think there still some points worth being discussed in this thread
<snip>
2. Make the review more transparent. If I understand correctly, almost anyone can be a review manager (criteria unclear) and that the acceptance is not decided counting yes and no votes. What happens however when the Review manager is not knowledgeable about the subject? How should he decide whether to accept or not? Authors want a fair treatment of their review.
<snip>
Marcelo
Boost Review Managers are vetted by the Review Wizards <https://www.boost.org/community/reviews.html#Wizard>. Part of the criteria is summed up as: "Approve the review manager based on initial acceptance by the library submitter, their participation in the Boost community, including the mailing list, previous reviews, and other forums." The goal is that the Review Manager is knowledgeable in the subject. There have been a few exceptions to that. If appropriate community participation occurs with experts, the process can still work fine. The Review Manager's duties are described here: <https://www.boost.org/community/reviews.html#Review_Manager>. The job of the Review Manager is explicitly not to tally up votes. They are to weigh the input from reviews and make an informed choice about the readiness of a library to enter Boost. Not all reviews are equal. Some reviews are very detailed and provide a great deal of background and reasoning. Other reviews are brief and look only at documentation or perhaps attempt to compile and run examples. All reviews provide important feedback but the feedback is used appropriately by the Review Manager. Boost reviews are somewhat grueling because of the intensity and level of expectation. Very often the Review Manager and Author will come to agreement for a negative outcome. The job of Review Manager is tough and requires some amount of finesse. Are you aware of a review in which author was unfairly treated? michael -- Michael Caisse
On Tue, 19 Apr 2022 at 20:27, Michael Caisse via Boost <boost@lists.boost.org> wrote:
<snip>
requires some amount of finesse. Are you aware of a review in which author was unfairly treated?
Thanks for the detailed information and links. I didn't want to suggest there has been an unfair review in Boost nor that it needs changes. I was speculating on the consequences of the rules as I knew them i.e. not counting up votes (from your link above) *The Review Manager: Decides if there is consensus to accept the library and if there are any conditions attached. Consensus is not the same as a vote. The Review Manager has discretion to weigh opinions based on authority or thoughtfulness.* Marcelo
participants (29)
-
Adam Wulkiewicz -
Alan de Freitas -
Andrea Bocci -
Andrey Semashev -
Bjorn Reese -
Dominique Devienne -
Gavin Lambert -
Glen Fernandes -
Ivan Matek -
John Maddock -
Kostas Savvidis -
Marcelo Zimbres Silva -
Marshall Clow -
Mateusz Loskot -
Matt Borland -
Michael Caisse -
Niall Douglas -
Peter Barker -
Peter Dimov -
Phil Endecott -
Rainer Deyke -
René Ferdinand Rivera Morell -
Richard Hodges -
Robert Ramey -
Stefan Seefeld -
Thomas Rodgers -
Vinnie Falco -
William Linkmeyer -
😉 Good Guy 😉