about venti: probably nobody got around to removing it. send a patch. more seriously: i have no clue what might be wrong with venti, cause i haven't used it for decades. the papers for venti and fossil are nice btw, i have nothing against the concept, rather i'm all for it, in theory.
On Wed, May 15, 2024 at 7:33 PM Lucio De Re <lucio.d...@gmail.com> wrote: > > Factually, Fossil is no big deal. Its design shortcomings have been raised in > the past from the Bell Labs side and the documentation (for Venti, I think, > but it's not very important) suggests that Fossil was knocked together as a > minimal Venti cache so the benefits of Venti could be utilised and the old > file system could be abandoned. > > I somehow missed that discussion at the time and never went back to find out > how it panned out. But it sure feels with hindsight that Fossil became the > trigger for the 9front schism and that would explain the sensitivity on > either side. It's a shame, because the Venti potential remains unrealised as > there isn't the Fossil bridge (where development is continuing, in 9front) to > a better, full functionality file system that includes Venti backing storage. > > Which brings me to the question I have been meaning to ask: what scope does > Venti serve in the absence of Fossil? I appreciate that VAC is a handy form > of archiving, but does it justify the complexity of configuring and > maintaining a Venti archive? I know that vacfs has some failings I haven't > had the opportunity or the inclination to investigate, but exhibit themselves > only in P9P - in my experience. So is Venti only a trophy application, or are > there serious uses for it among the 9front community? > > Lucio. > > On Wed, May 15, 2024 at 7:04 PM Jacob Moody <mo...@posixcafe.org> wrote: >> >> On 5/15/24 11:20, Don Bailey wrote: >> > >> > I have zero emotional attachment to Fossil. What I am asking for, not even >> > demanding, is a fact-based assessment of the asserted issue. Pointing at >> > the code is not an emotional attachment. It's literally the opposite. It's >> > asking to demonstrate and document the issues, instead of asserting that >> > something is awful because /you/ have had an emotional reaction to it >> > failing. How did it fail? Can you reproduce it? What code is bad? Why is >> > the code bad? If you can't answer these questions, maybe you >> > shouldn't have removed it. >> >> The emotional accusation I understand, it really seems like it's just fossil >> that evokes this >> reaction out of people. Just fossil that makes people want us to prove >> without any reason of doubt that the >> code should have been removed. I also just don't understand why people are >> so attached to fossil. >> Is it because people feel like there is a high burden of evidence for >> touching the holy code >> as ordained by bell labs? We didn't want it so it went. If you think this is >> actually a >> mistake and there is a world of possibility to be had thanks to fossil in >> Plan 9 I encourage >> you to maintain fossil yourself and prove to us that we were wrong in >> thinking it was dead weight. >> >> I want to specifically compare the discussion that happened on this thread >> between p9sk1 >> and fossil. We think that no one should be using p9sk1, and so we spent the >> time to explain >> to others the very real, concrete and specific issues with the code and >> implementation. >> >> We are not telling any other user of Plan 9 to not use fossil if they'd >> like, we simply don't want to >> deal with it in 9front. I think the burden of proof you are putting on us to >> make this >> decision would only make sense if we were advocating for other distributions >> and current >> users of fossil to no longer use it. It's fine, we're just not interested in >> it, sorry. >> >> As I, and others, have pointed out now a couple of times. Adding fossil back >> to 9front >> is trivial. Perhaps you haven't had the experience of having to sit in irc >> and help >> new users get going with the system who really don't have opinions about >> anything and >> then dealing with the outcomes when things blow up. As you said fossil is >> not exactly >> easy to deal with, it needs a lot of special consideration. So why then are >> you complaining >> that 9front made the decision to remove that option for the uninformed user? >> Does it not >> make more sense to direct users towards a filesystem that is more resilient >> and requires >> less watering? >> >> All of this is entirely moot with gefs right around the corner. I can't >> imagine someone >> willingly want to use fossil with gefs as a (soon to be) alternative. >> > > > > -- > Lucio De Re > 2 Piet Retief St > Kestell (Eastern Free State) > 9860 South Africa > > Ph.: +27 58 653 1433 > Cell: +27 83 251 5824 > 9fans / 9fans / see discussions + participants + delivery options Permalink ------------------------------------------ 9fans: 9fans Permalink: https://9fans.topicbox.com/groups/9fans/Tad3dc0c93039a7d2-M1843e75ffc3bf40e13ae37d0 Delivery options: https://9fans.topicbox.com/groups/9fans/subscription
