[seqfan] Re: A089755

[M. F. Hasler]

PS: IMHO it was not absolutely required to add a(1)=11 in the definition
(of course it does not harm to "spell this out", though)
because one can consider that this is implied by
"single digit primes are not allowed unless they arise from [a(n-1)
having 0 as 2nd digit]":
a(1) = 11 is the smallest prime compatible with this definition,
because for the first term a(n-1) is empty => has no 0 as 2nd digit =>
a single digit prime is not allowed.
-- 
Maximilian

On Sat, Sep 19, 2015 at 4:00 PM, M. F. Hasler <oeis at hasler.fr> wrote:
> David, did you have a look at the version
> https://oeis.org/history/view?seq=A089755&v=38
> I proposed since yesterday, a few hours before your message to seqfan ?
>
> On Sat, Sep 19, 2015 at 9:43 AM, David Wilson <davidwwilson at comcast.net> wrote:
>> Regarding A089755:
>>
>> I was finally able to cob together a program that almost works.
>>
>> Let "new" mean "not occurring previously in the sequence".
>>
>> Given element n, I cobbed together the following rule for computing the next element n':
>>
>> if (n is a single-digit number)
>> {
>>         n' = smallest new prime starting with n;
>> }
>> else if (next to last digit of n is 0)
>> {
>>         Remove leading digit of n;
>>         n' = smallest new prime starting with n;
>> }
>> else
>> {
>>         Remove leading digit of n;
>>         n' = smallest new prime > n starting with n;
>> }
>>
>> This rule is rather obscure and complicated, and is not deducible from the sequence description.
>> This supports my contention that the author was not clear about what he was doing.
>>
>> But even given the rule above, there are a couple of clear mistakes in the sequence.
>> By any reasonable definition we should have a(11) = 907 and a(20) = 701.
>> The sequence needs to have its elements fixed, or be deaded.
>>
>>> -----Original Message-----
>>> From: SeqFan [mailto:seqfan-bounces at list.seqfan.eu] On Behalf Of Frank
>>> Adams-Watters
>>> Sent: Saturday, September 19, 2015 1:29 AM
>>> To: seqfan at list.seqfan.eu
>>> Subject: [seqfan] Re: A089755
>>>
>>> David: no, the definition is consistent. You are not understanding what is
>>> meant by retaining leading zeros.
>>>
>>> After 13, the 1 is dropped, leaving 3. Since 1 digit numbers are prohibited, we
>>> can't get just make 3 the next term; it has to be 31.
>>>
>>> After 103, the 1 is dropped and we have 03, which is two digits and thus
>>> acceptable. This appears in the database as 3, because the OEIS doesn't allow
>>> leading zeros; but it's "really" 03.
>>>
>>> After 03, we drop the leading 0, and get something starting with 3: specifically
>>> 37.
>>>
>>> If I were to program it, which I probably won't, I would store the sequence as
>>> strings instead of as numbers.
>>>
>>> This sequence is clearly the work of someone who, at least at that time, did
>>> not understand how to use the empty string.
>>>
>>> Franklin T. Adams-Watters
>>>
>>> P.S. If someone can provide a better description, I'm fine with that.
>>>
>>> -----Original Message-----
>>> From: David Wilson <davidwwilson at comcast.net>
>>> To: 'Sequence Fanatics Discussion list' <seqfan at list.seqfan.eu>
>>> Sent: Sat, Sep 19, 2015 12:12 am
>>> Subject: [seqfan] Re: A089755
>>>
>>>
>>> A few notes on A089755 et al.(...)