## Not signed in

Want to take part in these discussions? Sign in if you have an account, or apply for one below

## Site Tag Cloud

Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.

• CommentRowNumber1.
• CommentAuthorAndrew Stacey
• CommentTimeMar 3rd 2009

I'm curious as to how far back the n-lab should go. I just created "constant morphism" but was debating whether or not that was too basic. I then found that Toby had edited it and made a Wikilink to "constant function". I'm curious as to what people think is the baseline for the n-lab.

I figure that something like "constant function" is a prerequisite for most of the n-lab. On the other hand, it might be an idea to have some "examples" pages showing how lots of the structure works for particular categories. Then "constant function" to link to the examples page for Set. This is a bit more pedagogical than perhaps we are at the moment.

• CommentRowNumber2.
• CommentAuthorTobyBartels
• CommentTimeMar 3rd 2009
This comment is invalid XHTML+MathML+SVG; displaying source. <div> To begin with, I don't think that anyone has to write <a href="http://ncatlab.org/nlab/show/constant+function">constant function</a> just because I made a link to it. But if (although in this case I find it unlikely) a lot of links to it start appearing, then that would be a sign that we really do need to write it.<p>If we did write it, of course, then we would have to write it as an example in category theory, not just as a definition in set theory. So we'd say how it matches the definition of constant morphism in Set (of course) and can also be defined as factoring through a terminal object, link to the pages on both those concepts, and refer to the concept of constant functor or any variations on that. In other words, we would want to mention its internalisations and categorifications, if anyone ever decides that its worth writing at all.<p>I don't have any doubts about <a href="http://ncatlab.org/nlab/show/constant+morphism">constant morphism</a> whatsoever. If you use the concept, then you need to explain what it means, particularly that it does <em>not</em> mean factoring through a terminal object. All category-theoretic concepts, no matter how basic, are valid pages in my opinion.</p></p> </div>
• CommentRowNumber3.
• CommentAuthorAndrew Stacey
• CommentTimeMar 4th 2009

Ah, I see. The 'constant function' proto-link was not a 'Someone needs to make a page on constant functions' but rather, 'if someone makes a page on constant functions, then this page ought to link to it.'

Is this usual Wiki-policy? It's certainly a reasonable point of view (and one that I'm fully prepared to work with - though I'd be tempted to scale down the intrusiveness of hanging Wiki-links if it is the general policy).

Your statement about 'all category-theoretic concepts' being fair game is pretty much what I was using as my working rule, hence my noticing the 'constant functions' link, I guess.

• CommentRowNumber4.
• CommentAuthorTobyBartels
• CommentTimeMar 5th 2009
In the old days on Wikipedia, we'd say that you should feel free to make links to pages that might not be written for a while, although there was also the attitude that everything is worth writing about in the end. (I don't know what the kids are saying there these days.) As for this wiki, standard policy is whatever we say it is, isn't it? I wouldn't link constant function from just any old page, but from constant morphism I would, even though I don't feel in any great hurry to write it.
• CommentRowNumber5.
• CommentAuthorAndrew Stacey
• CommentTimeMar 5th 2009

Seems reasonable enough. My only issue,then, is with how hanging links are displayed. I don't mind the question marks if they are temporary, but having something like: "a constant morphism in Set is a constant function?" makes it look as if we're not sure. "Bad form" as Captain Hook would say. Maybe I'm alone in this, but if not we could tweak the CSS so that hanging links looked like normal links except for the darker background. Then you click on the actual word to be taken to the create page link rather than the question mark. Or something.

• CommentRowNumber6.
• CommentAuthorTobyBartels
• CommentTimeMar 5th 2009
Yeah, Wikipedia switched from question marks to red links for the same reason. (But question marks remain an user-preferences option, especially for colour-blind readers.) This is done in CSS.
• CommentRowNumber7.
• CommentAuthorAndrew Stacey
• CommentTimeMar 6th 2009

I think a darker background solves the problem for colour-blind readers. Different coloured links can be mildly problematic as people can have default colours for links that they have and haven't followed so a followed-link can look like a non-existent link and vice versa.

It's possible to change the CSS on a local level, isn't it? I could try modifying Frolicher spaces to remove the question mark and see if anyone complains (or even notices!).

• CommentRowNumber8.
• CommentAuthorTobyBartels
• CommentTimeMar 6th 2009
Yeah, you can put CSS right there in the page source. Also, if you go to the web's HomePage and hit Edit Web and find Stylesheet Tweaks, you'll see what CSS is automatic on every page in that web only. And then http://ncatlab.org/stylesheets/instiki.css is the rest. Just so you know what all is in there.