Path: csiph.com!eternal-september.org!feeder.eternal-september.org!nntp.eternal-september.org!.POSTED!not-for-mail From: Tim Rentsch Newsgroups: comp.arch Subject: Re: OT writing direction Re: endian history Date: Tue, 15 Sep 2026 07:49:49 -0700 Organization: A noiseless patient Spider Lines: 96 Message-ID: <86tsnqvab6.fsf@linuxsc.com> References: <10uks4f$1dqo$1@gal.iecc.com> <10vnjp6$1df4$1@gal.iecc.com> <10vns1f$3a252$1@dont-email.me> <10vpblp$3mkd1$1@dont-email.me> <10vphh0$3o9jl$1@dont-email.me> <117gj6t$1b95v$2@dont-email.me> <6a9c1c93.39172234@news.eternal-september.org> <117h76c$1i80d$1@dont-email.me> <117hagr$1jf1n$1@dont-email.me> <117hc84$1k4o7$1@dont-email.me> <1788627455-5857@newsgrouper.org> <6a9cc83c.5722312@news.eternal-september.org> <117j391$24qgq$5@dont-email.me> <6a9db0f1.65298453@news.eternal-september.org> <1788739059-5857@newsgrouper.org> <117l0lh$2r1lf$1@dont-email.me> <117lfjn$2uuqs$1@dont-email.me> <86ik4gwvj0.fsf@linuxsc.com> <117nulc$3r0ij$1@dont-email.me> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Injection-Date: Tue, 15 Sep 2026 14:49:54 +0000 (UTC) Injection-Info: dont-email.me; logging-data="2667830"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1972fYREv0OR0Ms+sVpVqJK9sa9hyUYyGA="; posting-host="d22a44f9db2d80b4aae7023976fc11e4" User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.4 (gnu/linux) Cancel-Lock: sha1:i+2cJG0P5OLJCF0P8dJo8BByrdY= sha1:VKUlCNmzxuve98CYzFwQBBlSKd0= sha256:1asHeHoZQLtmT2QEAchi+pmfhQELeoCn0fpC7TwjrwI= sha1:IYQEXC0UmH/F/PLmeXnhJzftuIw= sha256:d3wzhQdXJ/ymDFeDb/vYO8gb8EmKyp1JPhKSbby5yCc= Xref: csiph.com comp.arch:118037 Stephen Fuld writes: > On 9/7/2026 3:23 PM, Tim Rentsch wrote: > >> Stephen Fuld writes: >> >>> On 9/6/2026 5:31 PM, Stephen Fuld wrote: >>> >>>> On 9/6/2026 4:57 PM, MitchAlsup wrote: >>>> >>>>> quadibloc@invalid.com (John Savard) posted: >>>>> >>>>>> On Sun, 6 Sep 2026 07:04:02 -0000 (UTC), Lawrence >>>>>> =?iso-8859-13?q?D=FFOliveiro?= wrote: >>>>>> >>>>>>> Endianness has nothing to do with writing direction. >>>>>> >>>>>> Not directly. >>>>> >>>>> Humans developed 3 writing orders {Left-Right, Right->left, and >>>>> Top->bottom}. I wonder why nobody standardized on bottom->top ?? >>>> >>>> Perhaps it has to do with the problems it would cause when trying to >>>> use long scrolls. :-) >>>> >>>> But ISTM that the choice of horizontal direction is independent of >>>> the choice of vertical direction, thus yielding four potential >>>> choices. >>> >>> Sorry for the self followup, but actually eight choices. For each of >>> the four corners, proceed in the horizontal direction or the vertical >>> direction. >> >> How about, for each of the four corners, proceed in a clockwise or >> anti-clockwise spiral? That's eight more choices. > > Now we are really going down the rabbit hole. There is an issue with > "spiral" directions. Let's say you start at the upper left corner, > then proceed to the right. When you hit the right edge, you are going > to go "down". But what is the orientation of the letters, i.e. should > you turn the paper 90 degrees to keep reading left to right, or should > the orientation of the letters change so you can now read top to > bottom without reorientating the paper? Both have problems. > > If you say you will change the orientation of the paper, what if the > "paper" is actually a bronze plaque mounted on a monument. Kinda hard > to change its orientation. And when you get to the bottom, do you > stand on your head to read it? :-) > > On the other hand, say you change the orientation of the letters. > That works for the right side - you just read it top to bottom. But > now what do you do when you get to the bottom? Do you print the > letters upside down, in which case you have the same problem as the > above solution. You could keep the letters "right side up:, but now > the words are spelled backwards. That's awkward. > > So I reject the spiral solutions, and other, even more bizarre > ones. But still, eight possibilities is more than enough. :-) I'm sorry you took my comments so seriously. They were very much meant tongue in cheek. That said, let me address your comments. If we are considering a Latin alphabet, which incidentally is a phonetic alphabet, the only ordering that really works is left-to-right then top-to-bottom. This follows from the shapes of different letters: they have different heights and widths, and also are stylized in a way to facilitate left-to-right scanning. The idea that the aforementioned eight possibilities are all viable doesn't hold up to serious scrutiny. If on the other hand we are considering a word-based alphabet, such as Chinese, things are very different. Characters typically have similarly sized bounding boxes, and are not stylized to prefer a particular reading direction: left-to-right, or right-to-left, or top-to-bottom -- all are workable. But even Chinese characters have a preferred up/down distinction, clearly favoring a top-to-bottom orientation over a bottom-to-top orientation. I'm pretty sure I have seen all of left-to-right, right-to-left, and top-to-bottom, for word-based alphabets, in regular texts. More exotic alphabets, as for example hieroglyphics or emoji or traffic signs, may occupy different categories. And of course squeezing all of these things into tiny pictograms on cell phones has distorted the notion of what we consider "language". (Side comment: some cell phones have the bad habit of spontaneously rotating from portrait mode to landscape mode, which I find thoroughly annoying.) In light of the above, IMO your eight possibilities are no more serious than the other possibilities I mentioned. It seems disingenuous to pretend otherwise. (My apologies for posting comments that have nothing to do with computer architecture.)