Bianchi modular forms -- beta?

[J E Cremona]

Since June we have had a couple of hundred thousand Bianchi modular
forms on the beta website, most linked to and from elliptic curves on
the production website. As I am about to submit a paper which
mentions some of these, for which I want to give URLs, it would be
really convenient for me to have this out of beta soon.

If you have not yet seen what is available, go to
http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/

Obviously if there are any serious outstanding issues I can fix those
before this happens.

John

--
Prof J E Cremona
Warwick Mathematics Institute
University of Warwick
http://www2.warwick.ac.uk/fac/sci/maths/people/staff/john_cremona/

[David Farmer]

The only concern I see is that URLs need to be maintained
once they appear in print.

The browsing page

http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/

has a perfectly fine name.

One could argue that the URLs of individual forms

http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/2.0.4.1/106.2/a/

would be better without the "ImaginaryQuadratic" part.
(And similarly for the "TotallyReal" in HMFs.)

It may be technically difficult to remove those from the URL,
because that would make it more difficult to figure out what
code handles the request. But I guess if we figure that out,
we would still maintain the old URLs.

So, looks good to me!

David

> --
> You received this message because you are subscribed to the Google Groups "lmdb" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to lmdb+uns...@googlegroups.com.
> To post to this group, send email to lm...@googlegroups.com.
> Visit this group at https://groups.google.com/group/lmdb.
> For more options, visit https://groups.google.com/d/optout.
>

[John Cremona]

On 12 September 2017 at 11:27, David Farmer <far...@aimath.org> wrote:
>
> The only concern I see is that URLs need to be maintained
> once they appear in print.

True. I'm already "translating" since the paper is joint with a
(graduated) student whose these has Bianchi labels (though not URLs)
from before the relabelling we did earlier this year (which is of
course the last ever relabelling).

>
> The browsing page
>
> http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/
>
> has a perfectly fine name.
>
> One could argue that the URLs of individual forms
>
> http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/2.0.4.1/106.2/a/
>
> would be better without the "ImaginaryQuadratic" part.
> (And similarly for the "TotallyReal" in HMFs.)

3 years ago I tried to unify the ImaginaryQuadratic and TotallyReal
lines but no-one was interested enough.

If we can get flask to behave I can see that we could make it so that

http://beta.lmfdb.org/ModularForm/GL2/I<field_label>

(where field_label could either be a label such as 2.0.4.1 or a
nickname such as Qsqrt5 or Q)

would map to the same with ImagainaryQuadratic/ inserted if the label
was 2.0.*.1, or with TotallyReal/ inserted if the label was d.d.*.*,
or with Q/ inserted if Q (which does not currently work, it should
perhaps redirect to /Q/holomorphic).

>
> It may be technically difficult to remove those from the URL,
> because that would make it more difficult to figure out what
> code handles the request. But I guess if we figure that out,
> we would still maintain the old URLs.

Yes, we would still allow the long versions.

>
> So, looks good to me!

Thanks -- I'll wait for some more feedback before making a pull request.

[John Voight]

I think I remember why we put "TotallyReal" and "ImaginaryQuadratic"
in: it was to 'group together' certain kinds of modular forms. I
don't see any harm in leaving it, except that it makes for a slightly
longer address, and I suppose that forms on miscellaneous fields won't
have a separate grouping, so that creates some inconsistency?

If I were to do this over again, I would keep the browsing pages as
http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/
etc., but delete it from forms, to be
http://beta.lmfdb.org/ModularForm/GL2/2.0.4.1/106.2/a/

But maybe we just want to leave it as is, since it's not broken.

I've played with the Bianchi forms a bit and I think it's time for
them to come out of beta.

JV

[John Cremona]

On 12 September 2017 at 15:17, John Voight <jvo...@gmail.com> wrote:
> I think I remember why we put "TotallyReal" and "ImaginaryQuadratic"
> in: it was to 'group together' certain kinds of modular forms. I
> don't see any harm in leaving it, except that it makes for a slightly
> longer address, and I suppose that forms on miscellaneous fields won't
> have a separate grouping, so that creates some inconsistency?
>
> If I were to do this over again, I would keep the browsing pages as
> http://beta.lmfdb.org/ModularForm/GL2/ImaginaryQuadratic/
> etc., but delete it from forms, to be
> http://beta.lmfdb.org/ModularForm/GL2/2.0.4.1/106.2/a/
>
> But maybe we just want to leave it as is, since it's not broken.
>
> I've played with the Bianchi forms a bit and I think it's time for
> them to come out of beta.

Thanks JV.

I don't know how we would deal with 2 URLS such as

http://beta.lmfdb.org/ModularForm/GL2/2.0.4.1/106.2/a/
http://beta.lmfdb.org/ModularForm/GL2/4.4.13068.1/holomorphic/4.4.13068.1-121.1-a

begin dealt with by different blueprints, but we could try having an
umbrella blueprint which would catch all ModularForm/GL2/ URLs which
were not caught by the more specific ones, and just serve the purpose
to look at the field label and decide who should deal with it. I
don't know if that is possible/

Looking at these 2 URLs (which of course don't work right now anyway)
I can see other discrepancies: The Bianchi one could say "harmonic"
to match "holomorphic"; and the last part is different -- but we have
an Issue for that already.

[John Voight]

> I don't know how we would deal with 2 URLS such as
>
> http://beta.lmfdb.org/ModularForm/GL2/2.0.4.1/106.2/a/
> http://beta.lmfdb.org/ModularForm/GL2/4.4.13068.1/holomorphic/4.4.13068.1-121.1-a
>
> begin dealt with by different blueprints, but we could try having an
> umbrella blueprint which would catch all ModularForm/GL2/ URLs which
> were not caught by the more specific ones, and just serve the purpose
> to look at the field label and decide who should deal with it. I
> don't know if that is possible/
>
> Looking at these 2 URLs (which of course don't work right now anyway)
> I can see other discrepancies: The Bianchi one could say "harmonic"
> to match "holomorphic"; and the last part is different -- but we have
> an Issue for that already.

There is something to be gained mathematically by taking things that
behave the same way and having that parallelism evident in the labels.
So in the two above, I would vote for:
http://beta.lmfdb.org/ModularForm/GL2/2.0.4.1/harmonic/106.2/a/
http://beta.lmfdb.org/ModularForm/GL2/4.4.13068.1/holomorphic/121.1/a/

Better we put this into place now than later? Maybe that's not so
crucial, because we can always redirect.

JV

[John Cremona]

I think it is easy to have the Bianchi URLs wirk with and without the
"harmonic" part. I'll try that. If it works then a similar trick
would also work for Hilbert.

>
> JV

[David Farmer]

I would have thought it is better to not use the
words "holomorphic" or "harmonic".

I agree with JV that if we had to do it again, we
would eliminate words like "TotallyReal" from
the URLs of individual forms.

So why not do that now, since we can?

Or at least, not insert the word "harmonic" where
it isn't really needed. If it will take some effort
to do it properly (i.e., make the word "holomorphic"
optional), then it is better to leave it inconsistent,
then to make it worse just to be consistent.

[Edgar Costa]

If you will indeed change the urls for HMFs, please consider avoiding the dashes:

if we ask for 
http://beta.lmfdb.org/ModularForm/GL2/TotallyReal/3.3.469.1/holomorphic/3.3.469.1-266.2-a

we should be redirected to:
http://beta.lmfdb.org/ModularForm/GL2/TotallyReal/3.3.469.1/holomorphic/266.2/a

I mentioned this in: https://github.com/LMFDB/lmfdb/issues/2202

Cheers,

Edgar

To unsubscribe from this group and stop receiving emails from it, send an email to lmdb+unsubscribe@googlegroups.com.

To post to this group, send email to lm...@googlegroups.com.
Visit this group at https://groups.google.com/group/lmdb.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "lmdb" group.

To unsubscribe from this group and stop receiving emails from it, send an email to lmdb+unsubscribe@googlegroups.com.

[John Cremona]

On 12 September 2017 at 15:57, Edgar Costa

<edgar...@math.dartmouth.edu> wrote:
> If you will indeed change the urls for HMFs, please consider avoiding the
> dashes:
> if we ask for
> http://beta.lmfdb.org/ModularForm/GL2/TotallyReal/3.3.469.1/holomorphic/3.3.469.1-266.2-a
> we should be redirected to:
> http://beta.lmfdb.org/ModularForm/GL2/TotallyReal/3.3.469.1/holomorphic/266.2/a
>
> I mentioned this in: https://github.com/LMFDB/lmfdb/issues/2202

Yes that's the issue I referred to earlier. It's concenient for the
full label (complete with dashes) to just work, but the official
version of the URL (as that would redirect to) should be the one with
/ as separators.

>>> email to lmdb+uns...@googlegroups.com.

>>> To post to this group, send email to lm...@googlegroups.com.
>>> Visit this group at https://groups.google.com/group/lmdb.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "lmdb" group.
>> To unsubscribe from this group and stop receiving emails from it, send an

>> email to lmdb+uns...@googlegroups.com.

>> To post to this group, send email to lm...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/lmdb.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "lmdb" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to lmdb+uns...@googlegroups.com.