|
Boost : |
From: Vinnie Falco (vinnie.falco_at_[hidden])
Date: 2024-07-10 21:33:47
This is a lovely thread, and I am hopeful that it will remain on-topic.
On Wed, Jul 10, 2024 at 7:48â¯AM David Sankel via Boost <
boost_at_[hidden]> wrote:
> For those who got involved in Boost within the last couple years, how did
> you hear about boost? What attracted you to it?
>
Interestingly, in 2001 when I wrote BearShare I tried out Boost.Asio as the
underlying network stack. My C++ was average back then, and I was using
Visual Studio 6. I got so flustered and turned off by the templates, the
concepts, compile errors going for days, that I just gave up. This sadly
left a bad taste in my mouth for Boost which carried on for several years.
When I came to Ripple, it was already using Boost.Asio through
"websocketpp." This websocket library was so frustrating for us that I
explored writing a new websocket library using Boost.Asio. My C++ had
gotten better and later I had access to more knowledgeable folks like
Howard Hinnant and Peter Dimov who I learned quite a lot from. I got the
idea to propose the library, renamed to Boost.Beast, which was subsequently
accepted. I suppose the motivators for Boost for me are:
1. Boost can connect to the internet (unlike std).
2. Boost helps me target earlier versions of C++, expanding my potential
audience.
3. The convenience and synergy from a release that is tested together and
works together.
For those who have been around for a while, what keeps you here? Why do you
> stay engaged?
>
I love the experience of having users. Lasting happiness is only achieved
in service to others; knowing that people out there benefit from my work,
whether commercially or otherwise, brings me joy. Most C++ developers
immediately recognize a great library when they use it, even if they
themselves do not quite rise to the level of skill required to author one.
A library is the equivalent of a work of art; I am happy to labor through
all aspects of development such as perfecting an API, writing
documentation, authoring tests, applying optimizations, creating
benchmarks, promoting the library to encourage its use, and supporting its
growing user community after it is published, as these skills are essential
for being well-rounded and the producing the greatest good.
Permissively licensed software is cheap. You write it once, at a fixed
price (time). The potential gain is infinite, as the acquisition of each
new user costs nothing. I stay engaged because of the emotional payoff at
the end of a development cycle when users get their hands on my newly
published stuff.
Thanks
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk