Path: csiph.com!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail From: "Carlos E.R." Newsgroups: comp.os.linux.misc Subject: Re: What Thinkest Thou Of LO Donate Banner? Date: Tue, 16 Sep 2025 21:42:47 +0200 Lines: 34 Message-ID: References: <1864d8e7ae136b94$115$2498948$802601b3@news.usenetexpress.com> <10a4ph3$obcm$2@dont-email.me> <68c5ef9f@news.ausics.net> <10a5sou$1360o$7@dont-email.me> <10a8gkd$25q4b$1@news1.tnib.de> <10a8i6f$1ppb4$2@dont-email.me> <10a8ju3$2a3hu$1@news1.tnib.de> <10a8kp1$1qcsq$1@dont-email.me> <10a8ob6$2ftau$1@news1.tnib.de> <10aa3vi$27nkf$2@dont-email.me> <10aavon$1covf$1@news1.tnib.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-Trace: individual.net 2v2xZoFsSp0hMFEgH1caCAsKEnAFsLCxXZALy4KXCS2Fum0cnI X-Orig-Path: Telcontar.valinor!not-for-mail Cancel-Lock: sha1:CFWNktDgaHEdpTWjWHw8Be9eSoc= sha256:+JXeKe/UJ0S8+sW2/FMD+UqAuuUzlw0ayaPjdXWnoRo= User-Agent: Mozilla Thunderbird Content-Language: es-ES, en-CA In-Reply-To: Xref: csiph.com comp.os.linux.misc:74343 On 2025-09-16 21:07, Richard Kettlewell wrote: > "Carlos E.R." writes: >> On 2025-09-16 09:53, Richard Kettlewell wrote: >>> Marc Haber writes: >>>> Lawrence D´Oliveiro wrote: >>>>> On Mon, 15 Sep 2025 12:04:21 +0200, Marc Haber wrote: >>> [large email services de-duplicating attachments internally] >>>>> You don’t see any of them offering such a dedupe service, do you? >>>> >>>> I'd expect that to happen transparently and invisible for the user. >>>> Again, server side. >>> Experimentally, I put a message with a 15MB attachment through Gmail >>> twice, the second time with the base64 encoding tampered with by >>> introducing some extra newlines. It preserved the tampered form >>> exactly. If it was extracting attachments for de-duplicated storage then >>> you’d expect to see the encoding canonicalized. >>> That doesn’t rule out that it de-duplicates the _encoded_ form of >>> attachments (or that there’s a threshold and it’s over 15MB...), but if >>> so then it seems like they’re leaving a lot of space savings unexploited >>> - you get a 25% saving just by storing the binary form instead of >>> base64. >> >> They can't do that. It nullifies signing, if it is employed. > > Fair point, although they could skip decode/re-encode compression in > signed messages only. Instead, they could just use compressed filesystems. In Linux, that means btrfs, it is the only one I know that supports transparent read-write compression. -- Cheers, Carlos. ES🇪🇸, EU🇪🇺;