[seqfan] Re: On Formulas

[franktaw at netscape.net]

Yes, the reason I'm bringing this up now, and didn't previously, is 
precisely because it isn't practical with the current system, but is 
with Wiki.

We need to be sure that we don't lock the format in too tightly.  In 
particular, there often are multiple formulas of each type, and 
certainly not all types need be present.

(One other problem with Maximilian's suggestion is that recursive 
formulas are not always best written as "a(n+1) = ..."; sometimes "a(n) 
= ..." is more convenient.)

One more type, that I omitted, is transforms.

Franklin T. Adams-Watters

-----Original Message-----
From: Maximilian Hasler <maximilian.hasler at gmail.com>

Yes, of course, on the wiki this will (easily) be possible via a 
template, e.g.
{{formula | ogf= ... | egf= ... | rec = ... | direct=... }}

once again, either authors respect this formatting directive,
or we can write a bot scanning for "improperly edited" sequences
(e.g. where the fromula is given as
===Formula===
* a(n) = ..
* o.g.f. =  ...
etc)
and doing the necessary changes automatically or semi-automatically.

Maximilian

On Thu, Dec 3, 2009 at 3:18 PM, Olivier Gerard 
<olivier.gerard at gmail.com> wrote:
> Dear Maximilian,
>
> I think you must consider this idea in the perspective of the OEIS 
Wiki.
>
> We can have in each relevant sequence article a formula section (this 
is
> already the case) but with
> a preformated model and a standard order, which can look like a table
> displaying typeset formulae for instance in front of formula types as
> suggested by Frank
> .
> This kind of approach is routinely used in parts of Wikipedia for 
things
> such as Atomic Elements, Countries, and other categories where
> multiple possible attributes are known and have to be collected in
> a systematic manner.
>
> For instance look at the table on the right of this article:
> http://en.wikipedia.org/wiki/Uranium
>
>
>
> Olivier
>
>
> On Thu, Dec 3, 2009 at 14:50, Maximilian Hasler
> <maximilian.hasler at gmail.com> wrote:
>> I agree that this would be useful, but maybe it cannot be done in the
>> current/ old system of "tags" (1-letter code), probably most among
>> a-z, A-Z are already used....
>> so the current (already "existing") system of "subdivision" is
>>
>> %F a(n) = ... (for direct formula)
>> %F a(n+1) = ... (for recursion)
>> %F o.g.f. = ...
>> %F e.g.f. = ...
>> %F Annn = { n | .... }
>>
>> The software displaying the entries could very easily do some
>> additional formatting (descriptive label in column 1, separate table
>> row...)
>> if authors respect that format.
>>
>> Maximilian
>>
>> On Thu, Dec 3, 2009 at 6:12 AM,  <franktaw at netscape.net> wrote:
>>> One thing I've thought would be nice for some time is to subdivide 
the
>>> formulas section.  What I'm thinking of is separate entries for 
direct
>>> computation, recursive computation, ordinary generating function,
>>> exponential generating function, Dirichlet generating function,
>>> asymptotic behavior, and other formulas.
>>>
>>> Franklin T. Adams-Watters