- From: Mark Davis ☕ <mark@macchiato.com>
- Date: Wed, 27 Apr 2011 11:57:06 -0700
- To: Shawn Steele <Shawn.Steele@microsoft.com>
- Cc: "Aharon (Vladimir) Lanin" <aharon@google.com>, Mohamed Mohie <MOHIEM@eg.ibm.com>, Matitiahu Allouche <matial@il.ibm.com>, "bidi@unicode.org" <bidi@unicode.org>, "bidi-bounce@unicode.org" <bidi-bounce@unicode.org>, "public-iri@w3.org" <public-iri@w3.org>
- Message-ID: <BANLkTin4rL5+jrDHeWZap9wU-76VhFLm0Q@mail.gmail.com>
For my mind, everyone is better off having a constant order over all IRIs. As Mati points out, the ordering of IRI fields is arbitrary anyway. It is * not* consistently big-endian; the most important part (the domain name) is little-endian. People get used to such inconsistencies as long as they are constant and predicatable. Mark *— Il meglio è l’inimico del bene —* On Wed, Apr 27, 2011 at 11:43, Shawn Steele <Shawn.Steele@microsoft.com>wrote: > That seems to be the biggest factor (locale), though I’m hesitant to > suggest it’s the only factor. > > > > For example, an Arabic speaking coworker using an en-US box obviously has > en-US locale preferences. However they can change their IE address bar to > “right to left reading order”. Then you’re suddenly outside of the “locale” > based assumptions. > > > > Also, “locale” in some cases would be more appropriately the page than the > user locale? Eg: visiting an en-US web page may have a different behavior > than an ar-EG web page? > > > > -Shawn > > > > *From:* Aharon (Vladimir) Lanin [mailto:aharon@google.com] > *Sent:* Wednesday, April 27, 2011 11:31 AM > *To:* Shawn Steele > *Cc:* Mohamed Mohie; Matitiahu Allouche; bidi@unicode.org; > bidi-bounce@unicode.org; Mark Davis ☕; public-iri@w3.org > *Subject:* Re: [bidi] Re: BIDI? > > > > So then we are talking about a locale property, not an (independent) user > setting. I could buy that. > > > > Aharon > > On Wed, Apr 27, 2011 at 9:16 PM, Shawn Steele <Shawn.Steele@microsoft.com> > wrote: > > Ø I think that the standard should be complete without relying on user > preferences. > > > > There seem to be clear differences in expectation, particularly between > Hebrew and Arabic speakers. It is possible to make a standard that is > consistent and doesn’t vary. However that standard might “look funny” to > some users. Worse, it seems that the preferences are biased by things like > how much of a mathematician they are. Probably all people going to > http://www.zoo.org aren’t engineers/scientists, and that needs to be > friendly to them. > > > > I think this is similar to date or time formatting. Yes, that’s “scary”, > because we’re talking about an IRI after all, which is Important. However > if I get a check for $1.000, the distinction between . and , is also going > to be important to me. > > > > So an IRI advertising a musical on the side of a bus may be a little fuzzy > to me if I don’t understand the culture. However I need to know the culture > to know whether to not bother typing in the IRI because 1/9/2011 has already > passed, or to make reservations because that’d be a fun thing to do on my > birthday. > > > > -Shawn > > > > >
Received on Wednesday, 27 April 2011 18:57:34 UTC