BLU Discuss list archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Discuss] CrowdStrike Fiasco (eBPF)
- Subject: [Discuss] CrowdStrike Fiasco (eBPF)
- From: daniel at syntheticblue.com (Daniel M Gessel)
- Date: Tue, 30 Jul 2024 18:13:37 -0400
- In-reply-to: <20240730165956.15f0eb41.Richard.Pieri@gmail.com>
- References: <87r0ba3dmk.fsf@hobgoblin.ariadne.com> <a8a4ed10-9dbc-4697-ba91-216699b7dd64@borg.org> <20240730165956.15f0eb41.Richard.Pieri@gmail.com>
I don't think embarrassment over poor quality metrics, like a high open bug count, are what keeps software in-house. Preserving the competitive advantage granted by a multi-million dollar investment was the main concern. Or not wanting to get bogged down providing support: creating good documentation is as hard as coding; "pull requests" need to be reviewed by senior developers. And everybody's got their opinion about naming conventions, indentation and where curly braces should go... A "conversation" that goes viral could be a massive time sink. And execs don't really want developers to "interface with the public". My problem with proprietary software is it is so often "consumer" oriented - the big "computer" operating systems have been narrowing focus to non-technical users, "managed desktops" making writing code more difficult. I do use consumer devices to watch TV or make phone calls (with the power of supercomputers from my early days) and that's fine, but it's just not want I want from my "computer". On 2024-07-30 16:59, Rich Pieri wrote: > On Tue, 30 Jul 2024 13:04:31 -0700 > Kent Borg <kentborg at borg.org> wrote: > >> But there is no reason proprietary software can't be good, just that >> that one pressure for quality is reduced when sources are kept secret. > I think this doesn't hold up to practical scrutiny. I have seen plenty > of horrible, ugly open source programs: everything Lennart Poetering has > written, for example. Or ShareLaTeX which was (might still be) so bad > that it won't compile outside the developers' unreproducible build > environment. > > In my experience, the license has nothing to do with quality. What > matters is people and time: enough people with appropriate skills and > sufficient time will produce something good. Cut any of these and the > results will not be good. Cut all of them and the results will be > worse. How many mission critical open source projects are supported by > too few people in their spare time? > > The only bearing the license has on this is that volunteer programmers > typically aren't forced by managers and publishers to do the impossible > to meet arbitrary release schedules. Of course, abusive management > practices are not required of proprietary software as Larian Studios > demonstrated with Baldur's Gate 3. So... yeah. License is not a measure > of software quality. >
- References:
- [Discuss] CrowdStrike Fiasco
- From: worley at alum.mit.edu (Dale R. Worley)
- [Discuss] CrowdStrike Fiasco (eBPF)
- From: kentborg at borg.org (Kent Borg)
- [Discuss] CrowdStrike Fiasco (eBPF)
- From: richard.pieri at gmail.com (Rich Pieri)
- [Discuss] CrowdStrike Fiasco
- Prev by Date: [Discuss] CrowdStrike Fiasco (eBPF)
- Previous by thread: [Discuss] CrowdStrike Fiasco (eBPF)
- Next by thread: [Discuss] CrowdStrike
- Index(es):