- Boost - lists.boost.org

Boost Asset Stewardship Review from Matt
by Matt Borland 11 Sep '24

11 Sep '24

DISCLAIMER: I am a C++Alliance Employee Prior to my employment by the C++Alliance I had been a contributor to Boost for a few years primarily in the Math domain. I enjoyed working with everyone, so when an opportunity to join the C++Alliance came around I took it. I agree with the foundation that the core of Boost is volunteers, but by adding a number of part time and full time employees we are able to handle much larger projects and sustained maintenance that may just be otherwise unfeasible. I don't think, for example, GCC would be better off without RedHat employees working on it because many people depend on it just like many people depend on Boost. I like that the C++Alliance is heavily invested in the future of Boost and it's infrastructure. Drone is fast and has a diversity of architectures that aren't available in Github Actions. We've been adding CUDA support to Boost.Math and a testing environment for that was happily provided since there is a heavy demand signal from our users. The new website is a pretty clear improvement over the current one. We also have Sam Darwin who is already helping the distribution of Boost happen behind the scenes. I see no reason to not want to work with an organization that literally puts their money where their mouth is. The fiscal sponsorship model and the new steering committee are the right choice for the future. The fiscal sponsorship model is not new and worked for Boost in the past. The three (possibly five) members of the initial steering committee all have public and sustained interactions with the Boost community. I trust they have the best interests of the project in mind since they have skin in the game. I would like to echo what Chris said in his review. Tensions have obviously been high over the past weeks and months between some rather large personalities. Once this referendum is complete I hope that we can have a peaceful transfer of power and return to a more cordial environment with a focus on how we can make Boost better moving forward. Bottom Line: I vote ACCEPT for the C++Alliance's Fiscal Sponsorship Proposal Matt

2 1

[Asset Stewardship] A request for clarification
by Andrzej Krzemienski 10 Sep '24

10 Sep '24

Hi All, I have a couple of questions to the C++ Alliance representatives regarding the Fiscal Sponsorship Proposal. First, the proposal is very clear and informative. Thank you for this. I still have some questions, though. In Appendix 1: Paragraph 2.c says, "all community programs [...] shall be the ultimate responsibility of Alliance. What is meant by "community programs" here? Paragraph 3 says "should Alliance be required to pay any taxes [...]". What taxes do you mean here? Can you give an example? Paragraph 3 also uses the term, "Project agrees". Is this a mistake, or intentional? I understand that the Current Committee can agree to something. Or that the Steering Committee can agree to something. But how can the Project agree? Additional two questions. 1. When I am a maintainer of a library, what does it make me in the light of this agreement? Am I part of the Project? Or a donor to the Project? Or something else? 2. I have a problem internalizing who decides and signs-off on the complement of the Current Committee. Don't get me wrong. I personally support the proposed complement (including the proposed amendments); I just wonder how it works logically/legally. The Current Committee cannot appoint itself. The Alliance cannot arbitrarily say "you, you, and you". Is this Boost Asset Stewardship Review a vehicle for the legitimization of the Current Committee? Paragraph 8.d says, "If no Successor is found, Alliance may dispose of the Project assets and liabilities in any manner consistent with applicable tax and charitable trust laws." This sounds very harsh to me. If the Project faces the necessity to find another sponsor, and this fails, is the Alliance allowed to offer the domain name for sale, get rid of the mailing list archives, or offer the logo to someone else? Appendix 3 -- I do not see how it illustrates the boost.org Domain Registration Expiration Appendix 6 -- It is very nice promotional material. Thank you for doing this. But the term "the Official C++ Language Slack Workspace" draws my attention, and I have always wanted to ask this: what makes this slack workspace "official"? Neither Boost nor the C++ Alliance represents C++ or could make decisions or blessings in the name of C++. So, is it not a stretch? Regards, &rzej;

3 5

Boost Asset Stewardship Review - Some questions received
by Glen Fernandes 10 Sep '24

10 Sep '24

The following questions were sent to me about the assets that the C++ Alliance already manages. My replies are included: 1. What are these alternate domains, and how do they differ from boost.io ? These are domains that the C++ Alliance provided for us when we had that issue where boost.org was down for a long period of time. At the time I wasn't sure we would even get boost.org back or what the timeline for recovery would be. I thought that we should provide a backup domain that users can be directed to until the issue is resolved. The C++ Alliance offered to help and registered boostlibraries.org and boostcpp.org that point to our website hosting. boost.io is the domain for people to preview the new website that the C++ Alliance built, developed by Fastly with content contributions from Boost users and developers. 2. Why do we need to pay to support downloads if the hosting is free? The storage costs are minimal but the bandwidth is not. At one point Boost downloads approached 60TB per month and Bintray couldn't afford to support us for free. JFrog generously offered to take over. Recently, our downloads started approaching almost 200TB per month, and now even JFrog couldn't afford to keep hosting us for free. They suggested we use a CDN to reduce traffic costs. I reached out to multiple providers to get a good deal. The best rate was from Fastly but this was also too expensive. The C++ Alliance offered to handle these costs and we subscribed to Fastly. Sam Darwin administers our Fastly account. For the list discussion, you can start at: https://lists.boost.org/Archives/boost/2024/01/255654.php Glen

4 4

Re: [boost] Replacing the ML with a forum
by Andrzej Krzemienski 09 Sep '24

09 Sep '24

śr., 26 kwi 2023 o 16:08 Vinnie Falco via Boost <boost(a)lists.boost.org> napisał(a): > Before we replace the mailing list with a forum (and implement a mail > gateway) can I get feedback on how you use the mailing list from your > email client? > Every message comes as a separate email to me. > > Are you using a threaded e-mail reader? Gmail groups all the emails from the topic as an email thread. > If so, do you care if > discussions are changed to be flat (like Discourse)? What does "flat" mean here? I am not familiar with Discourse. Could we get a sample or example? Regards, &rzej; What is important > to you (besides that you can continue to use your email client)? > > Thanks > > _______________________________________________ > Unsubscribe & other changes: > http://lists.boost.org/mailman/listinfo.cgi/boost >

5 7

Boost Asset Stewardship Review from Jakob
by Jakob Lövhall 09 Sep '24

09 Sep '24

I’m not affiliated with either the C++ Alliance or the Boost Foundation. I’m merely a boost user and sends occasional patches as contribution. The regulatory parts of the proposals are naturally US specific, and I want to make it clear that I do not understand the nuances of those parts. This is my short review. Boost Foundation: The bullet points and values stated in the proposal from the Boost Foundation seems strange to me and does not align with what I have been able to discern from public discourse. I haven't seen the transparency that’s being discussed. Engagement to provide said transparency seems lacking from the Boost Foundation. The emphasis on inclusivity over merit feels like a pathway to decline, as it explicitly excludes other groups beyond the intended inclusivity targets. This seems harmful to Boost and its community and should be avoided. The Boost foundation highlights code of conduct (and policing the discussion) as important going forward. I think it can be damaging, and should thus be avoided. Additionally, the Boost Foundation has not adequately addressed the infrastructure needs of Boost. In contrast, the C++ Alliance has stepped in to handle these concerns. C++ Alliance: Vinnie and others from the C++ Alliance have been actively engaging with the community on Slack and the mailing list—my two main points of contact. This provides transparency, builds relationships, and fosters discussions around technical issues, all of which are beneficial. The C++ Alliance has provided much-needed infrastructure assistance, particularly in cases where the Boost Foundation has failed to address long-standing issues (e.g., wowbagger). They support the maintenance (and at times creation) of libraries which I find useful. They promote on twitter, which I like, I've found good tools and libraries from twitter before and I expect others to also do so. having Boost on twitter therefore helps reaching those who might not have thought about boost when thinking about how to solve their current (C++) problems. The C++ Alliance focuses on merit over identity, which I believe is the right approach. Conclusion: I vote to ACCEPT C++ alliance proposal.

2 1

Regression Test Crashing
by Tom Kent 09 Sep '24

09 Sep '24

All- I've been pretty busy lately so just noticed this, but as of August 18 *all* the regression runs have broken during execution. The tail of the results log shows: - cxx11_char16_t : yes [47] - cxx11_char32_t : yes [47] - windows-sdk-supports-bcrypt : no [14] - windows-sdk-supports-bcrypt : no [51] error: Unable to find file or target named error: '/mnt/fs1/teeks99-06/run/boost_root/libs/charconv/build//stage' error: referred to from project at error: '../tools/check_build/test' Similar on windows: - cxx11_char32_t : yes [62] - windows-sdk-supports-bcrypt : yes [21] error: Unable to find file or target named error: '/C:/teeks99-09/run/boost_root/libs/charconv/build//stage' error: referred to from project at error: '../tools/check_build/test' Tom

4 8

Boost Asset Stewardship Review from Frank Wiles
by Glen Fernandes 08 Sep '24

08 Sep '24

Frank contributed the following review below, sent to me directly. > Vote: Transfer Boost Assets to the C++ Alliance > > Association with library: Occasional user. Long time OSS contributor > in other areas. Was paid to help C++ Alliance build the new boost.io > > Rationale: > > While my association with Boost is not large, I have seen first hand > the care the C++ Alliance has taken to both "do the right thing" and > drive for consensus with the Boost community. > > As someone who has been deeply involved in other software nonprofits > (Django Software Foundation directly, Python Software Foundation > tangentially) this proposal is certainly well thought out and > detailed. Most importantly it covers "end state" cases in the event > the Steering Council or C++ Alliance want to change or end the > relationship in the coming years. > > It seems clear to me that the C++ Alliance is driving the majority of > air in terms of breathing new life into Boost and the assets can > safely be within their control, if for no other reason than to reduce > friction in continuing to push things forward. > > -- > Frank Wiles > President/Founder of REVSYS > www.revsys.com > @fwiles Frank, thank you again for the review. Glen

1 0

Boost Asset Stewardship Review from Ali Asadpoor
by Glen Fernandes 07 Sep '24

07 Sep '24

Ali, thank you for the review. Everyone, Ali contributed the following review to glenfe -at- boost.org reposted verbatim here: On Fri, Sep 6, 2024 at 8:49 PM Ali Asadpoor wrote: > > Dear Glen, > > Hear is my review of the proposal doc: > > *** > Currently I don't see a bright future for boost library. Soon STL > will offer networking and later it might offer alternatives to other > awesome libraries in the boost. Seems STL is slowly catching up with > boost and it's killing it. > I was happy to read that The C++ Alliance might provide boost with > a second chance to glow again. Boost needs more human power to > process input from community faster and I think The C++ Alliance can > improve this aspect. > *** > > Best regards > Ali

2 1

Re: [boost] Boost Asset Stewardship Review begins 09/03 (Updated)
by Glen Fernandes 07 Sep '24

07 Sep '24

On Fri, Sep 6, 2024 at 7:44 PM Ion Gaztañaga wrote: > A: Yes, I think C++ Alliance’s proposal should be accepted, with a few > suggestions: Ion, a great many thanks for your review and the sentiments expressed within. Everyone, Ion's review didn't make it to the developer list because of an error in the address. I will ask him to repost it when he can tomorrow morning. Glen

1 0

Asset Stewardship Review
by M P 07 Sep '24

07 Sep '24

Here is my review of the proposed Boost.AssetStewardsipReview ## What is your evaluation of the design? The C++ Alliance has laid out a solid plan for boosting Boost, covering everything from revamping the website to improving integration, documentation, and library support. They’ve clearly identified the main issues and have offered practical, detailed solutions. They’re big on involving the Boost community and stakeholders, encouraging feedback and participation, and valuing everyone’s input. The Alliance is upfront about potential challenges like delays and costs, but they address these issues realistically. They’re not just offering quick fixes but are investing in new tools and dedicated staff for long-term support. On the flip side, the Boost Foundation’s concerns about the Alliance’s transparency and use of paid contributions might come across as overly critical. Sticking to slow decision-making and rigid governance could be seen as a barrier to innovation. Their focus on a purely volunteer-based setup might limit growth, and their critique of gender representation doesn't offer clear solutions. Some of their proposals, like improving mailing list behavior and adopting new tools, are a bit vague and might not be enough to attract fresh contributors or more dynamic approaches. ## What is your evaluation of the implementation? The implementation outlined in the text is well-organized and covers all the essential bases for Boost's long-term success. The C++ Alliance has already proven its value by setting up crucial infrastructure, like continuous integration and IT support, and by contributing to various libraries. The new Boost website is fantastic! Its modern design and user-friendly layout make it easy to navigate and find what you're looking for. With its fresh look and improved functionality, it's sure to catch the eye of new users and make their first experience with Boost a great one. Whether you're a seasoned developer or just getting started, the site’s clean interface and accessible resources are a huge plus. Their plan to revamp Boost’s brand and enhance its visibility is forward-thinking and aligns with what the community needs. The proposal to have the Alliance take on fiscal sponsorship aims to streamline operations and ensure financial stability. One thing to keep in mind is that the Alliance proposal comes with a legal agreement where a Steering Committee, made up entirely of Boost authors, gets to make decisions about the shared resources. If they're not happy with how things are going, they have the power to move those resources elsewhere. This kind of setup isn't mentioned in the Foundation proposal. So, while the idea of the Alliance handling fiscal sponsorship could definitely streamline operations and boost financial stability, how it plays out will depend on how well the Alliance and Boost community work together. ## What is your evaluation of the documentation? The C++ Alliance proposal is easy to follow and straightforward, clearly showcasing its contributions to Boost and laying out plans for future growth. It does a great job of explaining the benefits of the proposed governance changes and shows a real commitment to being open, sustainable, and collaborative. ## How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? I have dedicated considerable time to studying the proposals and related documents, and I have been aware of the ongoing conflict for some time. ## Are you knowledgeable about the problem domain? To some extent. With Boost's popularity declining in recent years, it's at a critical crossroads, and its future success may very well hinge on this decision. ## Do you think the C++ Alliance proposal should be accepted? Yes Miguel Portilla

2 1