[seqfan] Re: Concatenation

Charles Greathouse charles.greathouse at case.edu
Fri Apr 23 20:45:26 CEST 2010

Yes, precisely!  There are times when I want to search for something
but either the terms cannot be searched, or else there are too many
possible ways they may be written.

& was suggested to me as another possibility that is sometimes used
for concatenation.  So possibilities include &, +, //, U, ., a quoted
style, and concatenation itself.  I've seen & in some programming or
scripting languages; JavaScript, at least, uses +; Perl uses .; the
others I don't know.

Tangent: It's also effectively impossible to search for J programs, as
I found when I tried to see what obfuscated command it uses for that.

And thank you, Robert, for that comment on Maple.

Charles Greathouse
Case Western Reserve University

On Fri, Apr 23, 2010 at 2:06 PM, Rick Shepherd <rlshepherd2 at gmail.com> wrote:
> Good general(!) topic.  Either items like this should be standardized or
> else a complete, updated list of "acceptable" synonyms should be easily
> accessible from "search hints" pages -- and/or maybe "canned searches"
> (i.e., saved, predefined searches) could be provided.
> Some things to be aware of:
> Anyone interested in finding sequences related specifically to some form of
> concatenation but doing a naive search (without using the keyword option)
> would find that searching for
> 1) // or "//" is the same as searching for " " (all 174,000+ sequences are
> returned now),
> 2) concatenate or concatenated or concatenating will each find (different)
> sequences,
> 3) concat (trying to find all those above at once) would return sequences
> that concatenate arrays of terms in vectors using, for example, that
> PARI function (and not necessarily dealing with sequences whose terms
> themselves are the results of concatenation), and
> 4) concatenat finds no matches
> (there's also potentially even "str1" + "str2," not even mentioning again
> the other synonyms that you've already found)
> Presumably other restrictions would normally be placed on one's searches too
> but our goals should still include making searching as simple and intuitive
> as possible.  Standardization or maintaining lists of synonyms is a cost of
> that (or else relying on the individual users to find all the synonyms
> eventually themselves -- not that it's hard, just time-consuming).
> I know that the topic of concatenation is often not of interest as much to
> people with less interest in the keyword:base sequences but the general
> discussion about standardization of terminology and symbols to those *which
> can actually be searched for* easily is valid in any event.  (Perhaps with
> the switchover to wiki format there will be additional search techniques
> with wildcards, regular expressions, etc.? ...but these would probably be
> more useful to some people than others....)
> Would it be reasonable to consider the eventual development of some
> automated process to replace synonyms with the designated preferred synonym
> in each case (with viewing to check or for human intervention to override in
> unusual situations) before each submission (spell-check, grammar-check,
> terminology-check,...)?
> Regards,
> Rick
> On Fri, Apr 23, 2010 at 12:07 PM, Charles Greathouse <
> charles.greathouse at case.edu> wrote:
>> Sorry, I didn't mean to imply otherwise.  Essentially all of these
>> notations are defined before use, because none are common enough to be
>> known.  I wrote the original message as a result of coming across //
>> and not remembering having seen it before, even though it seems to be
>> the most common of the various notations I outlined.
>> Charles Greathouse
>> Analyst/Programmer
>> Case Western Reserve University
>>  On Fri, Apr 23, 2010 at 11:48 AM, Leroy Quet <q1qq2qqq3qqqq at yahoo.com>
>> wrote:
>> > It should be noted that I only sometimes use U (not my notation
>> originally). And even then, somewhere in the definition I always note that
>> this means concatenation.
>> >
>> > In my opinion, I don't care if there is a standard notation, just as long
>> as it is noted in the definition or comments that the notation used means
>> concatenation.
>> >
>> > Thanks,
>> > Leroy Quet
>> >
>> > [ ( [ ([( [ ( ([[o0Oo0Ooo0Oo(0)oO0ooO0oO0o]]) ) ] )]) ] ) ]
>> >
>> >
>> > --- On Fri, 4/23/10, Charles Greathouse <charles.greathouse at case.edu>
>> wrote:
>> >>...
>> >> Leroy Quet uses U for concatenation, as in A165784.

More information about the SeqFan mailing list