New lookup screen

[cino hilliard]

>From: "Russ Cox" <rsc at swtch.com>
>To: "cino hilliard" <hillcino368 at hotmail.com>
>CC: seqfan at ext.jussieu.fr
>Subject: Re: New lookup screen
>Date: Mon, 23 Jan 2006 11:37:07 -0500
>
> > Can we get a clear button to blank the look-up window?  Now the next
> > look-up
> > you have to back
> > out or highlight and delete out the prior entry in p4 windows xp.
>
>I don't know of a way to do this without JavaScript.
>I could take the old search out of the box in the results page,
>but I like having it there to edit to produce the next search.
>I often start with one search and then refine it.

Not a big deal. <Home><Shift><End><Delete> will do it key wise.
>
> > Also suggest we get the look-up field to accept multiple sequence
> > number
> > entries for example
> > id:A112798 A063376
> >
> > to bring up both sequences or several sequences.
> >
> > I read the hints twice.  Please forgive me if I missed it there.
>
>It's not there.  The search is a logical AND of the search terms.
>That's a pretty basic assumption built into the search mechanisms,
>sorry.  What would you use this for?  Why not just load two pages?

A112798 A063376 were reported by a fan to duplicates. I had to look each up 
in two browse
sessions to compare them. The feature would allow you to just scroll up and 
down.

A more important use (not that many dupes in the data base anymore :-)) 
would be to lookup
cross ref sequences in context you could simply copy
A064976 A062665 A028367 this_sequence A036298 A103169
then edit it to
id: A064976 A062665 A028367 A036298 A103169
copy and paste it into the panel.
This will engender a greater familiarity and understanding of the sequence 
and its sisters
with out the labor of individual lookup. Me, I may look up the first one or 
two and tire out from
the others. This may also make the browser faster since there will be 
scrolling on the PC vs server
requests by the persistent gotta know user.

>
>One could imagine using this for making pages listing, say, groups
>of related sequences, but you could just as easily write down the
>search terms for why they're related and use that.

If one requested a list of sequences with just one line of data  by the 
keyword "easy." a list
with 10 A numbers per line in the file would make the lookup easier by copy 
and paste 10 A numbers and then do the scroling and analysis on the PC.

Similar files could be create to list only sequences with a bibliographic 
entry in blocks of 10
with similar savings of server look-up.

Or a list of A numbers with the number first number > 1000,
etc



Cino