Talk:Limit (category theory)

This is the talk page for discussing improvements to the Limit (category theory) article.
This is not a forum for general discussion of the article's subject.

Put new text under old text. Click here to start a new topic.
New to Wikipedia? Welcome! Learn to edit; get help.

Article policies

Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

Untitled

I'm wondering if it would be better to separate out the stuff on colimits. In fact I've already started a colimit page. That way there would be more space to discuss the special cases such as products, equalisers and pushouts (and their duals on the colimit) page. Magnus 17:50 Apr 16, 2003 (UTC)

I think this article needs examples and motivations for the definitions instead of simply giving them out of the blue. Phys 18:08, 1 Sep 2003 (UTC)

Indeed! It also should include pointers to the most prominent limits/colimits and be more detailed and structured. Maybe I add something soon... -- Markus 25/11/2003

Done. But one may still think about a "motivations" section... --Markus 26/11/2003

limits in topology versus limits in category theory

Is there a way that you can interpret the limit of a sequence in a topological space as a limit of some appropriate functor between some categories? -Lethe | Talk 23:16, 8 October 2005 (UTC)

One would have to understand a topological space as a category. I don't have any good idea for this... -David 12/02/2006

Cones as natural transformations

Did you know cone may be views as natural transformations. It could be quite cool to explain limits with that point of view ? -David 12/02/2006

I've been meaning to incorporate that. I had stated some (minimal) notes at User:Fropuff/Draft 4. There is some more material at Universal morphism#Limits and colimits. -- Fropuff 20:18, 12 February 2006 (UTC)[reply]

Creation of limits

If have encountered in a paper the following definition of creation of limits:

G is said to create limits for F if whenever (L, φ) is a limit of GF there exists a unique limit (L′, φ′) of F such that G(L′, φ′) = (L, φ).

I consider this to be a rather weird definition of creation of limits, but the author of the paper insists on his definition. Is this definition used in a relevant portion of the literature? --Tillmo (talk) 14:49, 17 January 2008 (UTC)[reply]

Hi Till. Hmm. It's not Mac Lane's definition, but it's close enough to be confusing! Yes, it would be more tasteful if the author had used a word other than "creates". The most famous theorem about the creation of limits is probably Beck's monadicity theorem, and I guess that would fail for this weaker definition. Sam Staton (talk) 15:15, 17 January 2008 (UTC)[reply]

In the terminology used in this article G would be said to lift limits uniquely. This is indeed a useful concept, but not the one that is usually meant by create limits. Mac Lane doesn't talk about lifting of limits, only creation. The book by Adámek et. al. discusses both. (Although they use a slightly different definition of creation then that used here—which agrees with Mac Lane. They insist on a unique preimage source not just a unique preimage cone.) Note that creation of limits is equivalent to the unique lifting of limits plus reflection of limits. -- Fropuff (talk) 17:17, 17 January 2008 (UTC)[reply]

Definition of a cone seems strange. Please comment.

Er .. I miss something in the definition of a cone, which is allegedly supposed to generalize, among other things, cartesian products (think of J as a two-point discrete category): some property to assure that whenever we have a family (pair) of morphisms from an object in C to the diagram J \subset C (for simplicity), these factor through the cone N? Or (how) can one derive that from the functoriality of the diagram J -> C? For example, in a cartesian set product N = X \times Y, we want pairs of arrows A -> X, A -> Y to factor through N, so we need a morphism A -> N, but from this definition I dont see why it should exist. Will learn wiki markup later. Promise. H4nne (talk) 07:51, 30 April 2008 (UTC)[reply]

Taking J the two-point discrete category, a cone indeed is just a pair of arrows. A universal cone is a cone that factors through each (other) cone. You seem to expect that products = cones, but true is that products = universal cones. --Tillmo (talk) 15:17, 30 April 2008 (UTC)[reply]

Which diagram is depicted?

If the commutative diagram showing the cone is itself a diagram, is that diagram F:J→C, or another diagram in C? ᛭ LokiClock (talk) 21:55, 18 March 2012 (UTC)[reply]

I'm referring to the definition. ᛭ LokiClock (talk) 21:42, 19 April 2012 (UTC)[reply]

It's another diagram that contains part of the diagram F. Only part of the diagram F is shown, with nodes labelled FX and FY and the edge Ff. Hope that helps. ComputScientist (talk) 08:48, 20 April 2012 (UTC)[reply]

It does. My impression was a diagram doesn't contain the objects of the category, but the index category. So this suggests the reverse, that for an X in F:J→C there's a X' in the picture P:I→C such that P(X')=F(X). Then parts of a diagram are just subgraphs. Is that so? ᛭ LokiClock (talk) 11:15, 22 April 2012 (UTC)[reply]

Does it have to be, basically, based on Sets?

I mean, if you use Hom all throughout the article, it gives a wrong impression that categories in some way depend on sets. They don't; and Hom exists only for locally small categories. I'd rather rephrase everyting without using Hom.

Vlad Patryshev (talk) 18:19, 1 June 2012 (UTC)[reply]

It would be extra complication to add an extra C term to every mention of Hom throughout the article. Is there any general statement about categories made with Hom where it must be a set? ᛭ LokiClock (talk) 05:43, 3 June 2012 (UTC)[reply]

Change /varphi to /phi in the definition of the limit of a diagramm?

Right now, in the text it says (L, /varphi) and in the image, it says (L, /phi). Sorry for the LaTex-notation, I don't know how to write greek letters here. 134.76.62.225 (talk) 15:48, 11 March 2013 (UTC)[reply]

Backslash. Also, see Help:Displaying a formula. ᛭ LokiClock (talk) 18:47, 12 March 2013 (UTC)[reply]

New section (on limits in Set), and its tone

I think the effort that went into writing this new section is appreciated, however I think the tone is way too textbookish. But there may be some more fundamental problems with including it. I'm not sure the level of detail it goes into is really appropriate. We already have the general formula in terms of equalizer and product. Do we really need this too?

And statements like "since it is easier to speak of and think about graphs, we assume J is a graph with ..." are definitely inappropriate. Easier for whom? And it's not clear what you mean when you suddenly talk about taking a limit over a graph.

I think one or two explicit examples of the general formula for a couple standard cases (pullback, sequential, equalizer?) could be helpful, but what's here really just confuses things. Deacon Vorbis (talk) 15:50, 10 March 2017 (UTC)[reply]

Okay, after thinking about it some more, I'm removing the section -- it just has too many issues, but I do think a few explicit examples like I mentioned above would be good. Deacon Vorbis (talk) 18:46, 11 March 2017 (UTC)[reply]

> And it's not clear what you mean when you suddenly talk about taking a limit over a graph.

idk who said this, but one way to interpret this phrase is to think about the free/forgetful adjunction between graphs and categories (the free category over a graph being the one of all well-formed paths). there is a bijection between "pre-diagram"s, i.e. graph homomorphisms into the underlying graph of a given category from some "indexing graph", and diagrams in the same category indexed by the free category of the same indexing graph. so one might say the limit of a pre-diagram (graph homomorphism) is the limit of its corresponding diagram. 142.120.181.114 (talk) 05:39, 25 May 2022 (UTC)[reply]

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Limit (category theory). Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

Added archive https://web.archive.org/web/20081223001815/http://www.j-paine.org/ to http://www.j-paine.org/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 10:30, 23 December 2017 (UTC)[reply]

Possible typo in "Properties > As Representations of Functors” section

Towards the end of the section, I’m pretty sure the source of the natural isomorphism φ should be Hom(-, L), as it is an isomorphism between contravariant functors. For the same reason, the functor Cone(-, F) should go from C^{op} to Set, not from C to Set. 174.88.93.86 (talk) 19:56, 1 July 2024 (UTC)[reply]