search

schemedoc / awesome-scheme

A curated list of awesome Scheme libraries and resources
Creative Commons Zero v1.0 Universal
333 stars 24 forks source link

R6RS with/without errata #6

Open lassik opened 5 years ago

lassik commented 5 years ago

Current R6RS link is to http://www.r6rs.org/final/r6rs.pdf with a separate link to the errata listing at http://www.r6rs.org/r6rs-errata.html.

@weinholt You have errata-corrected PDFs up at https://weinholt.se/scheme/r6rs/. In my opinion we should forgo the separate errata listing and link directly to those. Or is there a particular reason someone would prefer to read the original erroneous versions instead? :)

You have the standard split into four separate PDFs (language; standard libraries; non-normative appendices; rationale). Should we link to at least the language and libraries?

arthurgleckler commented 5 years ago

On Fri, Jun 28, 2019 at 3:56 AM Lassi Kortela notifications@github.com wrote:

Current R6RS link is to http://www.r6rs.org/final/r6rs.pdf with a separate link to the errata listing at http://www.r6rs.org/r6rs-errata.html.

@weinholt https://github.com/weinholt You have errata-corrected PDFs up at https://weinholt.se/scheme/r6rs/. In my opinion we should forgo the separate errata listing and link directly to those. Or is there a particular reason someone would prefer to read the original erroneous versions instead? :)

While the corrected versions are good to have, they are not official R6RS standards documents. It's important to include links to the official standards and to distinguish between the two.

lassik commented 5 years ago

That's a good point. Is there a particular reason why the official standards committees do not distribute official errata-corrected PDFs, and could we ask them to do so?

lassik commented 5 years ago

Though in the absence of standards bodies like ISO, ANSI or IETF, the meaning of a "standard" not as strict. (Personally, I would argue that even in the case of those formal standards organizations, the implementations ultimately define the real standard.)

arthurgleckler commented 5 years ago

On Fri, Jun 28, 2019 at 8:45 AM Lassi Kortela notifications@github.com wrote:

That's a good point. Is there a particular reason why the official standards committees do not distribute official errata-corrected PDFs, and could we ask them to do so?

Frankly, because being on a standards committee is an enormous amount of work. The members are typically relieved to return to normal life after having finished and ratified a standard document. Putting out corrected versions would require reconvening and ratifying the revised documents, and there's no reason to believe that more errata will not arrive, requiring more work. It's easier to keep the errata separate and informally agree on them.

Also, since people, e.g. Scheme implementers, often do a lot of work based on standards documents once they have been issued, it's important for there to be essentially "diffs" between what they started their work based on and what's currently viewed as correct. The errata serve as those diffs.

lassik commented 5 years ago

Thanks for detailing the reasons. I'm not sure I whether I understand them but I accept them. If each batch of errata had to be blessed by the entire committee instead of 1-2 people then I definitely understand that nobody wants to do so much work.

lassik commented 5 years ago

PR #17 adds the R6RS docs