uned sequences (& preventing duplicate sequences follow-up)

[Frank Ellermann]

Rick Shepherd wrote:

> I have no problem with the editorial policy changing but
> consider the work involved in adding 40 sequences *twice*

Yes, that's not funny.  But the opposite is also possible, I
tried to edit a collection of "base" sequences, correcting and
adding a few terms, sent the result to Neil, and only one day
later I found that all these sequences were already listed in
the form a1(1), a1(2), etc., the new terms were simply defined
by a1'(n) = a1(n) * a1(n).

After this experience (i.e. deleting the corrected sequences,
because they were already listed in a much better form) I stay
away from "base" sequences, unless the %A mentions Patrick.

>> I consider potential "dupes" as a problem.
> I certainly agree with this.

If you want to check some suspicious sequences, I ran my dupe-
check again and published the results in 5 files:

<URL:http://purl.net/xyzzy/eis/eis_182.htm> 182..233 char.s match
<URL:http://purl.net/xyzzy/eis/eis_162.htm> 162..181 char.s
<URL:http://purl.net/xyzzy/eis/eis_161.htm> 161 characters, HUGE
<URL:http://purl.net/xyzzy/eis/eis_101.htm> 101..160 char.s
<URL:http://purl.net/xyzzy/eis/eis_100.htm> less than 100

The latter files are probably not really suspicious, and the
161 case is special (about 230 KB), but if more than 161 char.s
of terms in sequence A are also found in sequence B, then this
might a problem.  Or it could be documented with a reference -
the lists excluded references automatically (resp. I hope so).

> 1) If the editing workload becomes too heavy, a quota or
> throttling mechanism *could* be put in place so that the
> submitter him/herself must decide which is more worthy to go
> in (this day, week, month, year, etc).

LOL... this could be coupled with superseeker's limit (one per
hour), superseeker could send ticket numbers, and submissions
are required to have a superseeker ticket (e.g. the Message-ID)

> How big can "Offset" legally be?

AFAIK no limit, if you take 79 as line length, minus 11 for the
%O A567890 , then the remaining 68 characters could be split as
you like.  "60 digits ought to be enough for everybody" - but
not for the 1st signed term in http://purl.net/net/eisa/A057752

Okay, 60 digits might confuse superseeker, the linit is smaller,
but it's greater than 2001, see http://purl.net/net/eisa/A060958

>  Something like this would be more beneficial than seeing two
> "identical" entries that really aren't.

ACK, using three sequences with 1st = identical terms, 2nd =
more terms, 3rd = different terms is an interesting idea.  Who
tells superseeker about it ?  <gd&r>  That's no counterexample,
but some things won't work as expected with your proposal.  I'd
like to see these "continuation at first difference sequences",
and probably the number of these cases is very small.  So if a
lookup won't work directly there would be still references for
a manual lookup of the different continuations.

                       Bye, Frank