Registry fixed (was: Re: Records Missing Required Field: jkp, nph, tvt)

[Gordon P. Hemsley]

On Mon, Aug 27, 2012 at 10:56 PM, Doug Ewell <doug at ewellic.org> wrote:
> Gordon P. Hemsley wrote:
>
>> AFAICT, those additions were already in Registry version 2012-08-26.
>> The only problems were the 3 aforementioned entries (one of which was
>> a brand new subtag).
>
>
> This is true. The second 2012-08-27 version was the first one to have both
> the new and modified records AND the corrections to the clerical errors.
> That's all I was saying.

Ah, OK.

>> Everything is fixed now, but I worry: Is there any policy about how
>> many times the Registry is updated in one day? Because I already
>> downloaded the Registry version 2012-08-27 earlier, when it only
>> contained the delimiter correction. Now when I download it again, it
>> has those 3 entries fixed, too, but the date has not changed.
>>
>> Since the Registry is versioned by date, wouldn't it be best to ensure
>> that only one update occurs per day?
>
>
> Actually, Section 5.1 does say: "There SHALL be at most one version of the
> registry published in a day." I think the intent here was to cover
> well-formed releases of the Registry, not corrections to syntactically
> invalid releases, but the problem of stability based on date uniqueness
> exists either way.
>
> Strictly speaking, I suppose IANA should have waited until Tuesday to
> publish a Registry with the second set of corrections (the missing Added
> records), but the first two releases were ill-formed due to the errors and
> it didn't occur to me, when sending Amanda the corrections, that ill-formed
> releases were covered by this rule. I'll keep that in mind next time.

Yeah, because the ill-formedness of individual records does not, IMO,
invalid the entire Registry version. My parser issued an error on the
3 invalid records, but it had no problem parsing the remainder of the
file. It didn't even notice that there was an extra percent sign
somewhere. (In fact, I never even saw it.)

Since these particular errors were simply missing fields that were
deemed required, rather than other, more serious syntactic issues
(like errant or misused symbols, or encoding issues), the version of
the Registry should hold.

In this instance, the errors were localized to individual records,
which I don't think should invalid the entire Registry version. I
think, in the future, only issues that affect the parsing of the
entire registry should invalidate a Registry version.

Because if you can't parse the Registry at all, then you can't know
what version it is. ;)

Regards,
Gordon

-- 
Gordon P. Hemsley
me at gphemsley.org
http://gphemsley.org/ • http://gphemsley.org/blog/