Suggestion for the SageMath website

[Bill Hart]

Hi all,

I've been aware for a while that some libraries are not happy that their package is not explicitly mentioned by name on the sagemath.org home page and that you have to click through to the "and many more" to see their package credited.

I have a suggestion how this could be made more equitable for the many dependencies of Sage. The home page could randomly rotate the names of external dependencies that are listed on the main page so they all get equal exposure. It could even say, "Here is a selection updated randomly: ....... but there are many _more_". (My apologies if this is already the case, but after a number of refreshes, the list doesn't appear to change, presently.)

Obviously, the intention of this suggestion is to bring more prominent credit to authors of those other packages, which they deserve. As such, I'd prefer if we could avoid discussions to justify listing only certain packages on the main page (other than technical considerations, such as limited space on the main page), as this would have the opposite of the intended effect, which is to prominently recognize their work, which is well-deserved. 

I'm aware there may be other technical considerations, such as static vs dynamic nature of the html, SEO optimisation considerations related to regular changes to the page, etc. I guess those need to be considered. But I also believe this is a very important social issue, or I wouldn't raise it.

(The alternative, of course, would be to remove the short list on the main page and only have the very long list on the "and many more" page.)

Sorry it's taken me so long to get around to mentioning this, after it had been repeatedly brought to my attention by various people over the years.

Bill.

[kcrisman]

On Friday, June 1, 2018 at 8:46:52 AM UTC-4, Bill Hart wrote:

Hi all,

I've been aware for a while that some libraries are not happy that their package is not explicitly mentioned by name on the sagemath.org home page and that you have to click through to the "and many more" to see their package credited.

I have a suggestion how this could be made more equitable for the many dependencies of Sage. The home page could randomly rotate the names of external dependencies that are listed on the main page so they all get equal exposure. It could even say, "Here is a selection updated randomly: ....... but there are many _more_". (My apologies if this is already the case, but after a number of refreshes, the list doesn't appear to change, presently.)

This seems very reasonable; the list probably dates to a much earlier time when there were some "main" dependencies but now there are SO many ...  Can you make a ticket on http://github.com/sagemath/website so that Harald (and others who might have a way to deal with it) is more likely to see it?

[Bill Hart]

Done. [1]

[1] https://github.com/sagemath/website/issues/133

[Bill Hart]

On Friday, 1 June 2018 15:12:31 UTC+2, kcrisman wrote:

This seems very reasonable; the list probably dates to a much earlier time when there were some "main" dependencies but now there are SO many ...  Can you make a ticket on http://github.com/sagemath/website so that Harald (and others who might have a way to deal with it) is more likely to see it?

By the way, there are even some major dependencies not listed there. And these have been dependencies for many, many years. I accept that the list is there for historical reasons, but I don't think that is the reason.

As I say, I don't think it is justifiable, except on technical and historical grounds. The work of all of those people is obviously, and has obviously always been highly regarded and valued.

[William Stein]

PR's are very, very welcome!

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



--
William (http://wstein.org)

[William Stein]

On Fri, Jun 1, 2018 at 5:46 AM, 'Bill Hart' via sage-devel
<sage-...@googlegroups.com> wrote:
> I have a suggestion how this could be made more equitable for the many
> dependencies of Sage.

Regarding revamping the main landing page, i think it should be done
in a way in which all design decisions are focused around maximizing
value for our actual users. E.g., there should be a much more
prominent link to Gregory Bard's book "Sage for undergrads", and for
the French (and soon English) book that Paul Z. started -- since many,
many potential users of Sage are lost without those.

But further, linking to dependencies should be done much more, but in
a way that provides clear value to users:

- being able to better know what is in Sage
- being able to read the original upstreams docs and source code more easily
- knowing which upstreams devs to contact for *support*, to ask for
features, to contribute work, and to thank.
- being able to properly acknowledge what they are using

In any case, I hope whoever works on this project thinks about how
what they do will be used by users to make their experience with Sage
better overall, and benefit the community.

William

> The home page could randomly rotate the names of
> external dependencies that are listed on the main page so they all get equal
> exposure. It could even say, "Here is a selection updated randomly: .......
> but there are many _more_". (My apologies if this is already the case, but
> after a number of refreshes, the list doesn't appear to change, presently.)
>
> Obviously, the intention of this suggestion is to bring more prominent
> credit to authors of those other packages, which they deserve. As such, I'd
> prefer if we could avoid discussions to justify listing only certain
> packages on the main page (other than technical considerations, such as
> limited space on the main page), as this would have the opposite of the
> intended effect, which is to prominently recognize their work, which is
> well-deserved.
>
> I'm aware there may be other technical considerations, such as static vs
> dynamic nature of the html, SEO optimisation considerations related to
> regular changes to the page, etc. I guess those need to be considered. But I
> also believe this is a very important social issue, or I wouldn't raise it.
>
> (The alternative, of course, would be to remove the short list on the main
> page and only have the very long list on the "and many more" page.)
>
> Sorry it's taken me so long to get around to mentioning this, after it had
> been repeatedly brought to my attention by various people over the years.
>
> Bill.
>

[Emmanuel Charpentier]

Le vendredi 1 juin 2018 16:52:27 UTC+2, William a écrit :

[ Snip... ]

  E.g., there should be a much more
prominent link to Gregory Bard's book "Sage for undergrads", and for
the French (and soon English) book that Paul Z. started -- since many,
many potential users of Sage are lost without those.

Seconded. And I must add that the "Constructions" manual (which might benefit from some serious update and revamping) is also an underestimated and under-advertised resource when one is trying to get the "geist" of Sage... BTW, both our docs and tutorials and the above-mentioned books might also benefit of a few pointers to this under-ued resource.

But further, linking to dependencies should be done much more, but in
a way that provides clear value to users:

- being able to better know what Sage is 

  - being able to better know what is in Sage
  - being able to read the original upstreams docs and source code more easily
  - knowing which upstreams devs to contact for *support*, to ask for
features, to contribute work, and to thank.

The "normal" way to contact devs should be to use the mailing lists, the ask.sagemath.org site and (last but not least) the bug reporting system.

Here, I have to repeat that "normal" users (i. e. non-developers) should be able to file a ticket (or, at least, a ticket proposal) against Sage : reporting an (apparent) bug is a great way to contribute to software, by either exposing an hitherto unreported problem or exposing an unclear or unsufficient documentation...

[Bill Hart]

On Friday, 1 June 2018 16:48:14 UTC+2, William wrote:

>
> As I say, I don't think it is justifiable, except on technical and
> historical grounds. The work of all of those people is obviously, and has
> obviously always been highly regarded and valued.

PR's are very, very welcome!

I will not personally be able to work on this. 

I think you may have misunderstood what I wrote, though. I was referring to the work of all the people who wrote the libraries on which Sage relies. The Sage community has obviously always valued their work (independently of any PR's to the Sage project), and not being acknowledged on the front page doesn't represent anything except lack of space on the main page and the instantaneous recollection of the person who formulated that list. It's purely a historical and technical accident, it has nothing to do with how much their work is or isn't valued, hence my suggestion to randomise the shortlist or remove it. It appears to have the opposite effect of what is intended!

[Bill Hart]

On Friday, 1 June 2018 18:02:35 UTC+2, Bill Hart wrote:

<SNIP>  hence my suggestion to randomise the shortlist or remove it.

Another technical solution, if the list is retained in any form on the front page, would be to have a scrolling list. This might reduce the technical task to the insertion of a standard javascript element. Of course I know nothing about the technical requirements for the Sage website, e.g. load times, use of javascript, etc. It's just a technical suggestion, in case it is useful. 

[William Stein]

Great idea!

[TB]

On 01/06/18 17:51, William Stein wrote:
> But further, linking to dependencies should be done much more, but in
> a way that provides clear value to users:
>

> - being able to read the original upstreams docs and source code more easily

One idea in this direction: I usually use `function?` or `function??` in
the command line, but the HTML docs might benefit from using one of the
Sphinx extensions viewcode [1] and linkcode [2]. These extensions add
links next to functions and classes with the corresponding source code.

I do not know how easy will it be to add to Sage, because building the
docs is somewhat complicated.

Regards,
TB

[1] http://www.sphinx-doc.org/en/master/ext/viewcode.html
[2] http://www.sphinx-doc.org/en/master/ext/linkcode.html

[Simon King]

Am Freitag, 1. Juni 2018 21:04:00 UTC+2 schrieb mathzeta2:

On 01/06/18 17:51, William Stein wrote:
> But further, linking to dependencies should be done much more, but in
> a way that provides clear value to users:
>
>    - being able to read the original upstreams docs and source code more easily

 

Some Sage packages support building the docs. But how many do?

Would it be possible to have one Sage installation with all optional packages installed and building all available package docs, and post them along with the documentation of the Sage library? Say, have a prominent link to a list of all available packages and on that list have a link to the doc for each package that provides docs? Advantage: I am sure that this could be automatised.

Best regards,

Simon

[Vincent Klein]

Another alternative is to display all the packages in a slow scrolling horizontal list ScrollingPackagePrototype.html. 

Im not sure if the move is too disturbing for the reader or not. 

[Bill Hart]

On Wednesday, 6 June 2018 16:36:44 UTC+2, Vincent Klein wrote:

Another alternative is to display all the packages in a slow scrolling horizontal list ScrollingPackagePrototype.html. 

Im not sure if the move is too disturbing for the reader or not. 

That's actually exactly what I had in mind when I made my suggestion. I would personally put it on the lower part of the page, so that it is "below the fold" on most browsers (due to its scrolly nature). But maybe others prefer it higher up. 

[Vincent Klein]

Le mercredi 6 juin 2018 17:06:45 UTC+2, Bill Hart a écrit :

I would personally put it on the lower part of the page, so that it is "below the fold" on most browsers (due to its scrolly nature). But maybe others prefer it higher up. 

"below the fold", you mean under library and search button ? 

Side Note : The package list in the prototype is in alphabetical order, i keep in mind to randomize the list if we keep this solution.

[Bill Hart]

I just mean so it appears below the part visible in standard desktop browsers, i.e. so it is not the first thing you see when you load the page. "Below the fold" is my terminology for something that requires an action to read it, such as was required on this ancient IBM ThinkPad:

http://www.acontinuouslean.com/2009/10/01/the-original-ibm-thinkpad/

On side note: Is there any problem with alphabetical order if the whole list is there?

[Vincent Klein]

On side note: Is there any problem with alphabetical order if the whole list is there?

Yes in my opinion. The animation takes 180 sec to display the whole list. I think that's way longer than the average time people stay on the home page.

The result is a inequal exposure because the firsts in the list are the most viewed. 

[Bill Hart]

Yeah, I actually agree with that now after looking at it.

[Nikolay Hodyunya]

Many people would rather dislike this animation. As an alternative solution, why won't we just pick some random packages to show every time the page is requested?

--

[Vincent Klein]

Le dimanche 17 juin 2018 13:03:54 UTC+2, Nikolay Hodyunya a écrit :

Many people would rather dislike this animation. As an alternative solution, why won't we just pick some random packages to show every time the page is requested?

You may be right, but it's hard to decide as it's a matter of  personal taste (as always with UI design choice). Maybe should we do a poll ?

[Vincent Klein]

I have created a poll to help the decision https://framadate.org/kxRhiWt1dx95MCr5. It include the option to say the poll is a bad idea.

[Harald Schilly]

On Thursday, June 21, 2018 at 5:09:13 PM UTC+2, Vincent Klein wrote:

I have created a poll to help the decision https://framadate.org/kxRhiWt1dx95MCr5. It include the option to say the poll is a bad idea.

The poll has an option where it asks to add a comment here. My comment is to not show any packages on the landing page (neither random selected or some scrolling). Instead, a phrase like "consists of many packages", where "many packages" is a hyperlink to the component page. That component page is the alphabetically sorted list of all what's in sagemath: http://www.sagemath.org/links-components.html (and yes, I'm aware that the list needs work: https://github.com/sagemath/website/issues/135 ... contributions are welcome)

-- harald

 

[Peter Luschny]

> Instead, a phrase like "consists of many packages", where 

> "many packages" is a hyperlink to the component page. That

> component page is the alphabetically sorted list of all 

> what's in sagemath

Why not integrate this list into the page? That binds things

closer together. Like it is done here:

http://luschny.de/temp/SageHtmlTest.html

Peter

[Vincent Klein]

@Peter & @Harald, the poll has been updated with your proposal.

[Sébastien Labbé]

On Saturday, June 23, 2018 at 12:26:59 PM UTC+2, Harald Schilly wrote:

The poll has an option where it asks to add a comment here. My comment is to not show any packages on the landing page (neither random selected or some scrolling). Instead, a phrase like "consists of many packages", where "many packages" is a hyperlink to the component page.

My opinion is that the vast majority of people will not click on that link. On the contrary, with the list of few random packages, the vast majority will see some packages at each visit and each package will be shown a nonzero ratio of the number of visits.

I vote for the presence of a random sublist of the packages on the home page.

Sébastien

[Peter Luschny]

Sébastien Labbé:

My opinion is that the vast majority of people will not click on that link. On the contrary, with the list of few random packages, the vast majority will see some packages at each visit and each package will be shown a nonzero ratio of the number of visits.

I don't make any hypotheses about other people's behavior, 

how could I check that out? Instead, I will describe my own 

reaction:

In the case of a link (local or remote): If I see it for the 

first time, and have any interest whatsoever in the project, 

I will probably click it out of curiosity and read it with 

interest; later I wouldn't click it anymore unless I had a 

reason. I think that's fine for both sides.

In the case of a running banner: It annoys me the first time

I see it, and it will annoy me every later time I see it; 

I will not really look at it, I am used to displace such items

immediately from my field of vision just as I do this tens of 

times a day when I avoid commercial advertising trying to gain

my attention with similar techniques.

Peter