This site is supported by donations to The OEIS Foundation.

New OEIS Banner

From OeisWiki

Jump to: navigation, search

Contents

Contest For New Banner For OEIS

Report From the Judges

  • Here are the results of the competition for the new banner for the OEIS. Thanks to everyone who submitted a design - there were many excellent banners.
  • Rather than single out any one design as the unique winner, we have decided that the following designs all share the first place, and (at least for the first year or so) will be used in rotation.
    • D4a from Derek Orr
    • D7b from Susanne Wienand
    • D8 from Peter Luschny and the expanded versions similar to D8 which have illustrations, that is, D11 (Robert Munafo), D12 and D12a (Tilman Piesk), and D14b (M. F. Hasler). It is not clear right now which of these can be made to work with the present template that is used for the OEIS web pages.
    • D18a from Susanne Wienand
    • D21c from Philipp Emanuel Weidmann
    • D23 from Nick Devin
    • D27b from Arjun Jain
    • D28 (the static version) from Iakovos Ouranos
    • D29 from Pierre Cami
    • D30 from Pierre Lairez


  • Starting February 07, 2015, the banner will be Peter Luschny's design D8 again — this seemed to be one of the most popular of the designs we used in 2014.
  • Pierre Cami's design D29 was used from Nov 12 2014 to Feb 07 2015.
  • Iakovos Ouranos's design D28 was used from Sep 04 2014 to Nov 12 2014.
  • Nick Devin's design D23 was used from July 7,2014 to Sep 04, 2014.
  • Derek Orr's design D4a was used from May 15 2014 to July 7 2014.
  • Philipp Emanuel Weidmann's design D21c was used from April 1, 2014 to May 15, 2014.
  • Peter Luschny's design D8 was used from Feb 11 2014 to March 31 2014.
  • There will be a section in the main wiki page, https://oeis.org/wiki/Main_Page, pointing to this page, so that in the future you will always be able to see whose design is being used.
  • Incidentally, the Trustees of the OEIS Foundation voted 8 to 1 to include my name on the banner (I did not vote). Of course the database was not called the OEIS in 1964. From 1964 to 1973 it had no official name. In 1973 the Handbook of Integer Sequences (or HIS) was published, in 1995 the Encyclopedia of Integer Sequences (or EIS) was published, and the On-Line Encyclopedia of Integer Sequences (or OEIS) was launched in 1996.
  • Neil Sloane, Feb 26, 2014, with later additions and revisions

About the contest

The contest is now officially over.

  • The old banner, let's call this Design D0: Tony Noe pointed out that the banner at the top of the OEIS lookup page is out-of-date:
    File:IntegerSeq.gif
  • For one thing, it says "Integer Sequences" and it should say "The On-Line Encyclopedia of Integer Sequences®". It is also not particularly attractive.
  • We invite people to submit new designs by entering them in this page below.
  • The main requirement is that the wording should say The On-Line Encyclopedia of Integer Sequences®
  • The size of the banner is not too important right now — it will be adjusted later.

Detailed Instructions

  • To enter a design, log in to the OEIS wiki. Then click on the Upload file tab on the panel on the left to upload your design. Then click Edit (at the top of this page) and follow the source code used for Tony Noe's entry.
    • Here are more detailed instructions from Maximilian Hasler, Jan 11 2014:
      • Click on the [edit] tag to the right of the last section title (to avoid editing the page as a whole, but if that makes problems, click on edit at top of the page)
      • In the editor window, go right to the bottom of the contents (after the previous submission), enter (i) a headline and (ii) the line for the image and (iii) further comments as follows:
   == Submission by MyName, Jan XX 2014 ==
   [[File:the-name-I-want-to-give-to-my-image.png|center]]
   This is my submission, blabla...
   I put my name on top but sign it nonetheless: ~~~~
Then you preview that "post" (Alt+Shift+P). You will see a red link instead of the image (because it's not yet there) You can
  • save the post anyway (Alt+Shift+S), and add the picture afterwards as described below
or
  • add the picture as described below, preview & comment again, and safe afterwards.
To add the picture, you open the red link (with right-click ->> open in new window/tab, if you did not yet save the text). Then you will come to a page that explains how to upload the file, and it should be straightforward from there on (roughly, just as adding a b-file).
  • Comments on these designs are welcomed. First log in to the the OEIS wiki, then click Edit to leave a comment.
  • We will pick the winning designs in the next few weeks.
  • There will be no financial reward, but you will be thanked on the OEIS Foundation web site!
  • Submissions will become the property of the OEIS Foundation (like all other submissions to the OEIS)


  • Posted by Neil Sloane, President OEIS Foundation (email: njasloane@gmail.com), Jan 07 2014 - Feb 04 2014.

Design D1: A submission from Tony Noe, Jan 07, 2014

Comments

  • This incorporates the OEIS logo on the left, which is nice but is not a requirement.
  • As Tony said himself, the yellow in this banner is a bit too strong.
  • To see the banner in context, see Noe banner on sequence page and Noe banner on search page.
  • I made a wider and less-yellow version. Not sure how to upload it. Neil did the first one.

Design D2: A submission from Charles R Greathouse IV, Jan 07, 2014

Comments

  • This design is based on Tony Noe's 2014-01-07 banner.
  • It takes the color from the OEIS search bar, moves the text to a single line, keeps the yellow bar to a single line for lower visual impact, and adds a bleed to the logo. Because of the transparent areas above and below the text it can be set directly against text without additional spacing.
  • The primary purpose of this submission is to explore different visual ideas to inspire others; I (Charles) have no visual design experience.
  • But it needs "The" before "On-Line"! And if you are going to re-do it, could you use a more graceful font? The C and the Y look a bit clunky. Also the logo part is bit too big, perhaps?
    • Author note: The comment was added just as a new version with "The" was uploaded.

Design D3: A 2009 submission for the OEIS logo

Comments

  • This was submitted by a Sequence Fan back in 2009 when we were searching for a logo. It is not a candidate for the new banner, but is included here to illustrate a different kind of background.

Design D4: A Submission from Derek Orr, Jan 08, 2014

Comments

  • I took a different approach from the others...
  • On top, I have the sequence that started it all. In the background, I have numerous other sequences, all in the OEIS (specifically with the keyword "core"). I do apologize if I did not write the correct number in any of the sequences, I double-checked them to make sure I did.
  • Also, I kind of made my own OEIS logo on the side; I needed the logo to have no fill so I couldn't use the real image. I do prefer the original, however, without the fill. Without the fill, I think that would look really good. I am definitely open for any constructive criticism.
  • (From Neil Sloane) That's pretty nice! A couple of comments: For the sequence across the top, I would prefer the Catalan numbers (A000108): 1, 1, 2, 5, 14, 42, 132, 429, 1430, 4862, ..., simply because it the most important sequence after the primes. Whereas the sequence that started it all may be historically important, but as a sequence it is nothing special. Also, we definitely need a version of the logo (which has Recaman's sequence around the border, A005132, in red), with no fill. I'll ask David Applegate if he can produce that - he probably still has the original files we used for the logo.
    • (From Derek Orr) Awesome. I can definitely change the sequence on top. I agree; I believe the logo without the fill will look nice, too. I'll wait to do any concrete changes until David Applegate responds. Thanks!
    • (From Charles Greathouse) I've uploaded a transparent version of the OEIS logo here: File:Oeis logo trans.png. If David Applegate has a version which has higher resolution, so much the better.
    • (From David Applegate) I've uploaded a much higher resolution (2160x2160) transparent OEIS logo here: File:Oeis logo trans huge.png. The reason for the huge size is that (I believe) pixels in png files are either completely transparent or not at all. Since the smaller OEIS logo uses anti-aliased (or smoothed) characters, those partially red or black pixels would need partial transparency.
    • (From David Applegate) These logos were generated from the simple postscript file oeis_logo.ps, using ghostscript and netpbm. File:Oeis logo trans huge.png was generated by gs -q -dBATCH -dNOPAUSE -g4800x4800 -r416 -dBATCH -sOutputFile=- -sDEVICE=ppmraw oeis_logo.ps | pnmcut 320 816 2160 2160 | pnmtopng -transparent white > Oeis_logo_trans_huge.png. oeis_logo.png was generated by (something similar to) gs -q -dBATCH -dNOPAUSE -g1200x1200 -r104 -dBATCH -sOutputFile=- -sDEVICE=ppmraw oeis_logo.ps | pnmcut 80 204 540 540 | pnmtopng -transparent white | pngtopnm | pamscale -reduce 4 -linear | pnmtopng > oeis_logo.png
  • (From Olivier Gérard) Derek, if you change the sequence, try to put a little more space between the numbers and the title. They run a bit into each other in the current version. Also check if you can reduce the (r) mark and put it as an exponent of the last letter.
    • (From Derek Orr) I was going to do that, as well. I think the (R) would look better as a smaller exponent. I was also going to condense the title so the spacing between the two text lines is smaller. This should also satisfy Charles' recommendation below.
  • (From Charles Greathouse) I like the sequences in the background but not so much the one at the top -- it makes the logo very tall. (It's more of a big deal on mobile devices than on a monitor.) But that's just my view.
  • (From Derek Orr) I have another proposition. With the OEIS logo in full transparency, should I even have a central/top sequence that stands out from the rest? I could have only background sequences, which might be better. Since the OEIS logo includes Recaman's sequence, A005132, it might look less attractive with another main sequence. I'll play with it and see but any thoughts are appreciated.
  • (From Neil Sloane) Good point - let's try it without the sequence at the top, but with the no-fill logo. David Applegate says he can probably find a transparent version of the logo, but it might take him a little while.
    • (From David Applegate) I've now uploaded a transparent version of the logo at File:Oeis logo trans huge.png; the details are in my comment about 7 bullet points up.

Design D4a: Revision

Comments

  • I have uploaded the first revision of this banner. To me, I think it looks a thousand times better. I've done several things. First, I tried a sequence on top but it looked too confusing, like too much was going on in the banner. Also, by removing the top sequence, it shortens the banner size. I do have the sequence of Catalan numbers running off the banner on the bottom right. Further, I made the background sequences a bit less transparent so they are easier to read. I also removed some background sequences so the banner did not look too jumbled and hectic. In doing so, I rearranged them so there weren't too many empty spaces. I decreased the spacing between the two black text lines, made the (R) an exponent, changed the background color slightly, and added the transparent OEIS logo.
  • Further, if the background color were the color on the OEIS website (shown in Charles' submission), that might look better. Could anyone provide the color code so I might be able to test it out? Since I'm only changing the color, I won't make a separate Revision section; I would merely delete this banner above and replace it.
  • Derek, Marc LeBrun said the old background color was #ffffcc
    • (from Derek Orr) Thanks. I have uploaded the new banner with the new color scheme. I've also made it smaller.
  • (from Charles Greathouse) Derek, I think this is fantastic! Unfortunately the transparent version of the logo I gave you seems to be a bit jagged; hopefully you can get a better version which would be a nice (if small) improvement. Also there seems to be a 1-pixel gray line on the left which should probably be removed.
    • (from Derek Orr) I removed it, thanks. Yes it would be a small improvement but perhaps a crucial one.
  • Here's a picture of the old banner on the OEIS website.
  • (from Olivier Gérard) In the current revision, I think that the two lines of text are two high in the rectangle : the top of the first line's features are closer to the border than the bottom of the second line's descending elements.
    • (From Derek Orr) I'll try to fix that and make it look more aesthetic.
  • Maybe one of the sequences depicted on the banner should be something like A031214 or A102288, i.e. "OEIS of OEIS"?
    • (From Derek Orr) I like the fact that these are core sequences so people know what they are if they ever look at it. But this is a good suggestion, I encourage others to comment on this as well.

Sloane's Gap

  • (From Derek Orr) I already contributed a banner but I want to encourage the following: Perhaps a picture of Sloane's Gap should in the background. David Wilson's banner made me think of this. He has a picture on the right of the graph of Goldbach's Conjecture. To me, it looks really good. Thus, I believe having a graph in the background would be nice. Any thoughts on this?

Design D5: Nostalgia (by M. F. Hasler, Jan 8, 2014)

This is just a 0-th attempt of a banner that

  • conforms to the specs (wording),
  • retains the old font and colours,
  • includes the "new" OEIS logo (which I don't like too much, esp. for the "I" which should have serifs in order to get more "equilibrium", currently it looks (with the O above) like a magnifying glass -- well, which on a second thought is not that bad for a searchable database, in so far more as the remaining two letters "E[ncyclopedia of] S[equences]" still retain a significant part of the original meaning...) — M. F. Hasler 22:43, 8 January 2014 (UTC)
    • (From Charles Greathouse) I do think the magnifying glass is intentional. Actually, were I to redo the logo, I would make that more prominent.

PS: Wanted to add a comment, without completely subscribing to earlier (too radical) "chartjunk" discussion (although following the link to Ornament and crime was interesting to me). It is nonetheless universally agreed upon by specialists that documents should not have too much different fonts on the same page. Since the rest of the OEIS.org pages is in serif font, maybe the banner should not be in sans-serif font, although that may be considered these days (where people think that in 10 years they found out to know everything better than what their ancestors established during centuries or more) as more elegant.

Design D6: Colors Patterns (A submission from Shivam. N. Patel)

In this design I have done something quite different from the rest . Firstly the background is quite different - ie. A table of numbers. Secondly there is a Fibonacci series with the size proportional to their magnitude. Thirdly in The word - The online Encyclopedia of Integer sequences . The colors are set in such a way that the letter at prime place has different color (ie , green ) and that at composite place has different color (ie . Mauve ) Fourthly Vibrant colors are used in this design .

Comments

  • (From Neil Sloane) Very nice! Good to see a totally new approach.
  • The background image is a photo by Ivars Peterson. Do you have his permission to use it? —Brian Kell 07:00, 9 January 2014 (UTC)

Minor comment

  • The prev reminded me how the "on-line" bugs me a bit, though there might be some history for the dash unbeknownst to me. I did have the wacky thought that somehow something formulaic might be incorporated into the word string (like "e","i" etc). Too much probably--Bill McEachen 12:14, 9 January 2014 (UTC)
  • (From Neil Sloane) I don't understand your comment. Perhaps you are suggesting that "On-Line" should be changed to "Online"? There are several reasons why we don't want to do that. We have used "On-Line" since 1996. "On-Line" is the spelling in the trademark that we registered. "On-Line" is in my opinion the preferred spelling. "On-Line" is the spelling used in Webster's Ninth New Collegiate Dictionary. It is true that now (in 2014) "Online" is perhaps more common, but that's a pity.
    • (from Charles Greathouse) I come to the same conclusion by a different path. To me "on-line" is already archaic, almost phased entirely out of use. But I like holding onto it as a symbol of antiquity: so many websites come and go, but we've been around in more-or-less the same form since the beginning of the WWW, and with origins going further back of course. So to me it's a badge of honor that we use the old spelling.

Design D7: A banner based on A000002 (a submission from Susanne Wienand, Jan 9, 2014)

The red numbers around the banner are inspired by the current, very nice OEIS logo.
They show A000002 up to n = 59, moving around the banner. Starting point is the left upper corner of the banner. Three points above the number in the lower left corner indicate that the sequence is not finished with that number.
The squares in the background show the same sequence up to n = 84. Orange squares stand for a one, grey squares for a two. They start in the left upper corner and move around the banner spirally in the same direction as the red numbers. The second square in the third row represents the 84'th number. The sequence goes on further, but this is not indicated.

  • (From Neil Sloane) Excellent design! Great color scheme. And the Kolokoski sequence (A000002) is a good choice. A minor comment - "OEIS" should also have the little R-in-a-circle symbol, since that is also registered as a trademark.

Design D7a: Revisions

An '®' is added to 'OEIS'

As suggested by Neil Sloane, I marked OEIS as a trademark.

  • (From Michel Marcus) I like the colors, and the fact that it reminds me of a punched card.

Design D7b: 'founded in 1964 by N. J. A. Sloane' is added

I picked up Peter Luschny's idea to mention Neil Sloane. (Edit Susanne Wienand 17:06, 11 February 2014 (UTC))
Because of the additional text, in this banner there are 68 red numbers and 105 squares. Square no. 105 lies in the third row and the 19'th column.
I'm not sure about the starting year. In OEIS:Brief History, 1965 is mentioned as year of start.

(From Neil Sloane, Jan 13 2014) This morning I looked at the notebooks that I kept when I was working on my thesis at Cornell. They are labeled Rosenblatt I through V (Frank Rosenblatt was my thesis chairman). The initial collections of sequences are in volume II, which dates it to early 1964. I will change the Brief History page accordingly.

I uploaded a new version of D7b and replaced 'started' by 'founded', 'Neil' by 'N.' Susanne Wienand 17:16, 11 February 2014 (UTC)

Design D8: A submission from Peter Luschny, Jan 09, 2014

Comments

  • The banner is based on LaTeX. So it is formally described, easily reproducible and portable. At the same time it uses free high quality fonts. The design idea is more oriented at the New Sobriety (Neue Sachlichkeit) than on baroque illustrations of scientific information. The implementation is encapsulated in the following 6 lines LaTeX:
\input{OEISstyle.sty} 
\begin{document} 
\oeisBanner{{\hspace{-3pt}{The On--Line Encyclopedia of Integer Sequences\hspace{-4pt} \textsuperscript{\textregistered}}}}
\vspace{-18pt}
\raggedleft{{\fontsize{8pt}{10pt}\selectfont founded by Neil J. A. Sloane}}
\end{document} 
  • No additional logo or art design was used, deliberately; in the first place because the current design of the logo does not convince me (to me it looks like a Kindergeburtstagskuchen with a tennis racket). Instead I feel that it is more than appropriate to mention the creator of the database which (according to Wikipedia) is also cited simply as Sloane's.
  • The banner suggests some small changes in the format of the current header of the sequence pages: the line "Greetings from The On-Line Encyclopedia of Integer Sequences!" below the search box repeats the banner. Therefore I suggest to replace it by a link to the portal near to the 'Hints' link. Second, in order not to clutter up the header, I think the sentence "This site is supported by donations to the OEIS Foundation" is better placed in the footer. I do not think that this will decrease the number of donations. Finally the search box could be placed in the right upper corner, like in many other web sites including Wikipedia.
  • To see the banner in this context, see the banner on sequence page.
  • (Added later) Here is how it might look in the main OEIS page. Here is another example of how it would look.
  • (from Charles Greathouse) Very classy! I like your banner. I also think your suggestions are good, except that I like the search bar where it is.
  • (From Neil Sloane) I agree with Charles. Very classy indeed. I also prefer the search bar where it is now (beneath the search bar). Shouldn't the "f" in "founded" be capitalized? Answer: I guess not - I agree with Charles's comment below.
  • (From Al) If it was up to me, this one would be it, with one or two tiny tweaks. I would suggest "Founded by N. J. A. Sloane" because I think this form is more prevalent in bibliographies.
    • (from Charles Greathouse) Lowercase "f" is a bit less intimidating here, and it also works in the context of "The OEIS, founded by njas" (you wouldn't write "The OEIS, Founded by njas").
      I agree. I'm not sure but I think Al meant to change mainly "Neil" to "N.", not the lowercase. I also think it should be a little smaller than on the "example pages", in particular a little smaller than in the File:OEIS-Banner-Contest.png where it extends over the full width of the page. (What I don't like/understand is the LaTeX code shown: first, that code does not implement the banner, essential information is hidden in the .sty file and the obscure command \oeisBanner. Also, I don't think that using "hand made spacing" is advisable, and if so, it should be done using, e.g., \\[-.5ex] or the like, rather than \vspace{-18pt} [argh!]. Spacings and font selection should be relative to the initial size or at least the documentstyle's standard sizes (i.e., use \huge, \small, etc.)) — M. F. Hasler 04:25, 11 January 2014 (UTC)
  • (from Antti Karttunen) Notwithstanding my penchant for baroque illustrations, I like this one the best. Just wondering: would it be against the rules of typography if the two lines were centrally justified? Now it seems maybe a bit too asymmetric? Also, for my eyes, "Founded by N. J. A. Sloane" looks better than "founded by njas". We should also remember that OEIS is intended to be maintained for centuries, so in my opinion, it never needs to follow all the latest trends in web-design, but might well have a solemn and a bit conservative look.
    • (from AK) OK, "founded by ..." looks also good.
  • (from Olivier Gérard) I agree with Maximilian remarks on the TeX source. The idea is sound but one must be careful in the way it is encoded as this banner if chosen could allow to produce localized versions of the OEIS home page. Also, without adding too much baroque, and compromising its simplicity too much, there should be a little 'math' or 'science' element somewhere, probably in color.
  • (from Peter Luschny) Design 8a: This is an alternative, prompted by Olivier Gérard's comment "there should be a little 'math' or 'science' element somewhere, probably in color" plus Michel Marcus' remark on the 1973 book cover which leads to A001011. It might be a little bit too large for the banner but as a poster or as the dust jacket for the third edition it might fit. So I leave it here.
  • (From Enrique Pérez): Design 8 is the one I like the most, because as I use to print on paper, and keep in folders, several related sequences together with my own notes, about some topics of my interest, that's because I prefer to have everything in black and white, this logo saves expensive ink and looks nice at the top of the sequence, it looks useful to separate one sequence from another. Another advantage is that can be reproduced in latex. I agree to include Neil's name, I use to have many ideas everyday, the more simple the ideas are, the more difficult is to be rewarded for them. When you tell your simple idea to other people, at the first time they doesn't understand you, and afterwards they think that you are "highly" stupid, but the day after they sell the idea as theirs. To create a sequence of sequences looks simple, and even stupid, but the real thing is that it is brilliant.

Design D9: A submission from Noam Youngerman, Jan 09, 2014

  • The bit on the left can also be made into a square logo / favicon
  • (From Neil Sloane) Excellent design and the two shades of green are very pleasant to look at. The banner looks even better in your "In Context" page.

Design D10: From Alonso del Arte 16:28, 10 January 2014 (UTC)

  • Just for the sake of completeness. It's too drab, but at least, if people forgot the meaning of the background, they might be able to figure it out by finding the ratio of the rectangles in the rectangular grid that aligns the dots (roughly rationalized sqrt(2) to 1).
  • (from Michel Marcus, Jan 12) Pastel colors are nice. I like the forward movement given by the ES.

Design D11: From Robert Munafo 2014 Jan 11 03:49 UTC

I am a child of the Martin Gardner generation, when for example the MAA's current "Icosahedron" logo with its crosshatch shading was about as complex as logos ever get. I also think that figures like those in N.J.A. Sloane's original book, or those on the 2009 OEIS Wiki launch poster, capture the "joy" of the OEIS, and without figures it's not going to inspire those people that need a logo to be inspired.

Thus, I believe Peter Luschny's design is the best because it is monochrome, but it also needs some classic Martin Gardner-style line art to bring the "joy". I have added three figures from the 2009 poster to represent this idea. I think figures from the 1973 or 1994 book would be better, or others that NJAS thinks have been the most popular over the many years of the project.

Robert Munafo 04:11, 11 January 2014 (UTC)

  • (From Neil Sloane, Jan 11 2014): I like Robert Munafo's modification of Peter Luschny's banner and its choice of illustrations: trees, polyominoes and the graph of the EKG sequence. These are three quite different types of image, which is good.
    • I looked through the illustrations in the 1973 and 1995 books, but I didn't come up with any better images.
    • At first I was going to suggest replacing the EKG graph with the "toothpick" image (found in the top right corner of the OEIS poster), but that is another counting sequence, whereas the EKG sequence comes from number theory and is also a good representative of the 235000 graphs in the OEIS.
  • (From Michel Marcus, Jan 12 2014): Yes, nice, very classical. The 1973 book cover had golden folded stamps on a blue background.
  • (From Peter Luschny, Jan 12 2014) @Michel: Yeah, a nice collection of paper-clips on the cover there. Does anyone knows what they mean? Answer from N. J. A. Sloane: See Folding a strip of stamps, A001011.
  • (from Charles Greathouse) I like the banner, but the box around the rightmost illustration pulls the eye a bit too much for me. Are you attached to this?
  • (from Robert Munafo 20:07, 13 January 2014 (UTC)) I see now that I had been conflating the present question (a banner) with the broader issue of logo/brand/trademark/corporate-identity. Banners include logos, so it's related, but here we're just talking about banners. Also: I am not specifically attached to any individual image(s), as I stated above, the three I picked are just to represent the idea. Any of those in the Piesk examples below are good too, but I assert that it's more important that they be simple, non-distracting, and eclectic (e.g., don't have all 3 images be combinatoric). The mouseover-text and random images ideas are nice too, but possibly could be deferred since their implementation would delay addressing the original problem.
    • (from Charles Greathouse) I like all three that you've picked, with the small caveat above.

Design D12: From Tilman Piesk 2014 Jan 11

Banner

Robert's version above inspired me to add my 2 cents. The details are:

Responses

  • (From Michel Marcus, Jan 12) With minor case letters, it leaves more room for graphics.
I think this is a noteworthy remark. But it inspires me yet another one: IMHO we should (or at least could) dissociate two notions of "banners": The "full text trademark", and the "complete web page banner with ornaments and decoration". I think we should (try to) reach distinctly an agreement
  • about the exact way to display the registered/trademarked text (font, size, capitalization, aligment)
  • of how to decorate it: with our without a frame around, background colour, maybe animated (i.e., changing) illustrations, ... to the left and/or to the right of the text. — M. F. Hasler 14:20, 12 January 2014 (UTC)


Slider with random images

However, what I would actually like to propose is something different:
Let's drop the banner completely or choose a plain text version.
Instead just put a row of 3+ random images there, of the kind that Robert and I have put in our logos. They could be randomly chosen on page load from a set of 100+ images.
No description is visible, but when the mouse is moved over the image a mouseover like the following appears:

Α000105(4) = 5
The 5 polyominos with 4 cells (tetrominos)

And when the image is clicked it links to the sequence.
Images like the following could be used - often in a simplified version suitable for the small format:

There could be a small icon in the mouseover, that links to a description of the image itself.

Design D12a: How the slider could look:

Responses

  • Robert Munafo 18:53, 12 January 2014 (UTC) : The "random images with mouseover and click" idea is a great one, because it doubles as a way for visual people to explore and learn about the OEIS and all it has to offer. The idea reminds me of the WolframAlpha home screen (did you notice all those little tiny drawings are clickable?
Michel Marcus's point is also more in the direction I'd like to go. I think the text is the only essential part. It is persistent and timeless, and the graphics typically follow fashion trend of the decade. WolframAlpha and many maths institutions have a single simple graphic, which is treated as timeless, but if there are multiple or many images, as in my suggestion above, it's better the text be clearly separated from the rest.
  • Martin Y. Champel 26 January 2014 : I find the design very nice and clear, it reminds me the type of messages used in the Voyager Nasa program.

Design D13: From David Wilson 2014 Jan 11

  • Prominent, legible, central title with trademark symbol.
  • I think the banner has a good shape and size.
  • Acknowledgment of Neil per recent discussions.
  • Hopefully attractive color scheme.
  • Background interesting but not attention grabbing.
  • Background attempts to indicate function of OEIS.
  • Picked an interesting and mathematically significant sequence as background subject.
  • (from Michel Marcus, Jan 12) I like it, it has everything: terms, formula, graph, color, border. Could we see it in context?
I like it but I think it has too much numbers on the left. IMHO, one (or even 2) line(s) less would be "more". — M. F. Hasler 14:12, 12 January 2014 (UTC)
The sequence shows 100 elements, only 2 more than appear in the sequence. I think larger text might detract from the title. However, I can fix minor issues if they are a problem. David W. Wilson 02:33, 21 January 2014 (UTC)

Design D14: From M. F. Hasler, Jan 12 2014

The above is produced by the following 5-liner (complete LaTeX file, no fuzz):

\documentclass{article}\begin{document}\hrule
\begin{center}\huge\textsc{The On-Line Encyclopedia\\
of Integer Sequences}\,\raisebox{1ex}{\small\textregistered}
\end{center}\hrule\medskip
\raggedleft{founded in 1964 by N. J. A. Sloane}\end{document}

Design D14a: Changing the \textsc to \textbf yields the following:

These could be surrounded to the right and left by animated gifs or changing illustrations with further gadgets (mouseover...) as suggested by Piesk, above.

M. F. Hasler 16:12, 12 January 2014 (UTC)

  • (From Susanne Wienand) The starting year seems not to be clear. In OEIS:Brief History, 1965 is mentioned as year of start.
  • (From Neil Sloane, Jan 13 2014) This morning I looked at the notebooks that I kept when I was working on my thesis at Cornell. They are labeled Rosenblatt I through V (Frank Rosenblatt was my thesis chairman). The initial collections of sequences are in volume II, which dates it to early 1964. I will change the Brief History page accordingly.

Design D14b: graphics added

Upon Neil's suggestion, here's a Luschny-Munafo-Hasler mashup: (Looks blurred because scaled down here.)

Memo: Click the banner twice to see a clearer version.

Variants are possible (no upper line, line(s) going throughout the full width, "founded by..." completely flushed to the right. (Maybe the right side is too heavy w.r.t. the left side, and ornaments should be permuted, and/or another graphics (one of those chosen by Piesk) added to the left?) — M. F. Hasler 23:55, 14 January 2014 (UTC)

PS: actually, it can also be done in 4 lines of plain HTML, see User:M. F. Hasler/OEIS Banner :-) — M. F. Hasler 00:50, 15 January 2014 (UTC)

Design D15: Submission from James R. Buddenhagen, Jan 13 2014

This is my submission. James R. Buddenhagen 18:13, 13 January 2014 (UTC)

Design D16: Submission by IK, Jan 14 2014

this is my submission

  • (From Derek Orr) I'm not an expert on sequences but do the numbers in the banner represent any sequence or decimal of some sort? I like the idea!
  • (Martin Y. Champel) I also like the idea and the sobriety. I suggest you create a sequence based on your design : 7,0,2,3,1,5 just to answer "Yes" to Derek Orr
  • (IK) In fact OEIS gives 3 sequences that contain 7,0,2,3,1,5: A159811, A021911 and A145436.

Design D17: Submission from Mihir Mathur, 11 January

I reworked the OEIS logo, such that it has some elements from the old logo. The logo is quite mathematical; the top edge of the square has the prime numbers, the bottom edge has the fibonacci numbers, the right edge has the digits of the golden ratio and the left edge has the digits of pi. Furthermore, the sides of the outer and inner square are in golden ratio. As of now, the background is simple and neat. It can be made more "mathematical" if needed.

Design D18: Submission from Susanne Wienand, Jan 14 2014

This is a banner with a photo in the background.
Some integers are added to the photo in order to count some objects on it. Susanne Wienand 22:19, 14 January 2014 (UTC)
I took the photo at Phoenix-See, Dortmund. Susanne Wienand 04:53, 16 January 2014 (UTC)

Beautiful photo! - Charles R Greathouse IV 00:01, 21 January 2014 (UTC)

Design D18a: Numbers, their positions and their colors are changed

The surface of the clear lake invites you to let stones skip on it.
So in this revision, I added numbers of A136259, stone skipping numbers, to the photo. Read from left to right, they are the first 18 numbers of the sequence.
By the banner I want to express that in the world which surrounds us, relations to integer sequences can be found everywhere. Susanne Wienand 19:54, 16 January 2014 (UTC)

Comments: (from Martin Y. Champel): I share the importance to express the idea that integers were used from the earliest ages of mankind and existing even earlier.

"stone skipping numbers" in Google returns only 3 pages, all on OEIS. Where is i_2? Autymn Castleton 22:47, 29 January 2014 (UTC)
The digit '3', corresponding to index 2, is on the upper left side, above the 'h'. Susanne Wienand 15:05, 30 January 2014 (UTC)

I uploaded a new version of D18a: 'started' is replaced by 'founded', 'Neil' is replaced by 'N.', there are slight changes concerning positions and colors of the numbers. Susanne Wienand 17:13, 11 February 2014 (UTC)

Designs D19a, D19b, D19c, D19d: Submissions for the OEIS logo from Peter Luschny, Jan 15 2014

The third logo uses a design of Stefan Friedrich Birkner which is CCA-3.

Comments

  • Numbers are dull. Therefore I reject the idea to paint them on an OEIS logo or on the banner of OEIS pages.
  • What makes OEIS valuable for me is that it is a treasure chest of mathematical ideas and references which lead to the literature discussing these ideas. This is what a logo for OEIS should try to capture.
  • A word of explanation: In abstract set theory the set of integer sequences is often denoted by Z^N.
  • This is a really neat idea. These letters could also be incorporated into any other design. (However, Z and N should be blackboard bold.) Tilman Piesk 00:26, 16 January 2014 (UTC)
  • Tilman, yes, and indeed they are! I experimented with smaller sizes and found that the logo is better readable when the letters are filled. Compare with the earlier version which I added now. By the way, the background is the cover of the [Sachsenspiegel]. Greetings to Leipzig! Peter Luschny 08:51, 16 January 2014 (UTC)
  • You might place the 'Sachsenspiegel' on a shelf with books, in order to point to the literature referenced by the encyclopedia. Susanne Wienand 09:11, 19 January 2014 (UTC)
  • I too love the idea of , and even though we're discussing a banner, not a new logo, perhaps the banner might inspire a future logo change) Robert Munafo 06:09, 26 January 2014 (UTC)
  • A bit late to comment, but the second one from the left is really great! — Antti Karttunen 20:39, 1 March 2014 (UTC)

Design D20: Submission from Pierre CAMI, Jan 17 2014

Makes me think of the NFPA 704 diamond for some reason. If we choose this design we should also correct the initials -- "Alexander" rather than "aLexander". - Charles R Greathouse IV 00:04, 21 January 2014 (UTC)

Design D20a: Another version

Design D21: Submission by Philipp Emanuel Weidmann, Jan 22 2014

Main proposal
"Credits" version
"Credits" version with updated text, as suggested by Neil Sloane
Demo of banner in context
  • Integer sequences are as beautiful and diverse as the colors of the rainbow. In the OEIS, they all come together.
    • The design attempts to capture that idea.
  • Sequences shown are (as suggested by the welcome page and Derek Orr):
  • Designed with Inkscape (SVG available) using openly licensed fonts:
    • Raleway (Logotype, 5 different font faces used)
    • Linux Monospace (sequences)
  • As shown, the banner easily fits the results page, but it could also be made a little smaller vertically by "bending" the sequence lanes inward if desired.
  • By virtue of it being a typography-only design, the proposal also lends itself to printing use. - Philipp Emanuel Weidmann 07:49, 22 January 2014 (UTC)
  • Update: I did some tests with low-quality monitors (such as those that are - unfortunately - often used in university departments where OEIS is frequently accessed from) and based on that adjusted the yellow tone to give a nicer appearance with equal luminosity on such monitors, while leaving the appearance on top-quality monitors good. - Philipp Emanuel Weidmann 18:38, 22 January 2014 (UTC)
  • I really like this design. I definitely think that once a banner is chosen, it should be one that displays some sequences of some sort. And nice sequences chosen. Perhaps Triangular numbers should be one? Maybe instead of Mersenne primes. Then all the numbers seen could be around the same order of magnitude. - Derek Orr 05:08, 23 January 2014 (UTC)
    • Update 2: Thank you for the feedback. I've just uploaded new versions with various improvements to the design, including better font hinting and kerning, reduced height, and, as you suggested, the triangular numbers replacing the Mersenne primes for a more balanced layout (to anyone who has looked at the previous design: the aspect ratio has changed; please refresh your browser cache to see this). I'm quite pleased with the result. - Philipp Emanuel Weidmann 18:10, 23 January 2014 (UTC)
  • Great ideas here. If we're going to have a bunch of numbers (representing sequences) in the banner at all, this is the way to do it: the numbers form a graphic element, which is well-defined clear and simple and limited to a logo-like size and shape, and the digits can be ignored by the eye looking at just the overall graphic form. Similarly, I think this is the way to do color, if we're going to have color at all: the color is bright and interesting, but there isn't a lot of it, and only a tiny bit of information is lost if the colors were all changed to black. And the use of font-weights is very appealing too. Robert Munafo 06:24, 26 January 2014 (UTC)
  • I also like the idea & appearance. But shouldn't the sequences start from the logo and "go on / away to infinity"? (I understand that this would make the first (less interesting(?)) terms be printed in large and the more interesting larger terms in small, but conceptually I think it would be more logical. (Unless we can explain that the infinite rest of the seq. is "eaten up"/"inside" the OEIS?) Also, I don't understand why "Encycl." & "Sequences" get that much more weight than "on-line" and "integer"? — M. F. Hasler 02:54, 3 February 2014 (UTC)
    • The idea is that the encyclopedia (symbolized by the logotype) is the continuation of the sequences (as it hosts the complete versions of them, or a formula/explanation), so I think both the placement and the size makes sense with that in mind. As for the font weight, that is a purely aesthetic choice, enhancing the visual flow from left to right. - Philipp Emanuel Weidmann 06:51, 3 February 2014 (UTC)
  • Update 3: New version with credits wording as suggested by Neil Sloane. - Philipp Emanuel Weidmann 17:08, 11 February 2014 (UTC)

Design D22: Submission from David Z Crookes, Jan 22 2014

Herewith a submission for the logo, based on a copyright-free painting of my own.

Design D23: Submission from Nick Devin, 24 January 2014

Here's my design. The background is a snapshot of some digits of e. Also included is a sample page with the logo edited in.

Another version:

Design D24: Submission from Tim Lewis, January 24, 2014

This animated gif file was done in Matlab. It would not be too hard to create similar files with a sequence of the day (assuming the animation is not too gimmicky of an idea).

  • A little too distracting for me, but maybe if you slow it down to about 1/3 or 1/4 of the speed shown here :-), it would be fine. Robert Munafo 06:27, 26 January 2014 (UTC)

Design D25: Submission from Andy Nicol, 25 January 2014

Updated with a new font & "founded in 1964 by N.J.A Sloane". All images in the carousel are now <10k.

Some comments on implementation: I think it's a good idea to use several of these banners - there are some nice ones here. But if this carousel banner is chosen as one, it would require some additional work as it is not a simple image based header. My main worry is server load - loading three new images at random for each load of the page is a lot of extra overhead. I think three carousel images would have to be chosen at random to be static for a day/week or so - that way they could be cached for the user's whole session. As long as your web server supports gzip compression (for automatically serving a compressed version of the jquery libs) I think this banner could come in at <40k for first load, everything should be cached after that. Each time the carousel buttons were used it would load an extra 5k-10k image. I'm happy to do the extra javascript work if required, but I'd need input on how to proceed and suggestions/submissions for further images & links to add to the carousel.

(updated 12-2-14 by Andy Nicol)


This submission follows the comments about having a scrolling carousel banner to provide visualisations of the sequences. I think sites like Wolfram have shown that this type of layout is beneficial to expanding knowledge and encouraging exploration.

The header typefaces are now the open licence font Crimson Text. The fonts in the logo - Scriptania and ag stencil - are also open licence.

I've implemented a (hopefully) working version of the carousel here: http://www.planetside.pwp.blueyonder.co.uk/oeisbanner/oeis.htm

The banner chooses a random start point and loads three images. Hovering over the image brings up text with an explanation of the visualisation. Clicking an image forwards to the relevant page. Buttons on the side of the carousel allow the user to scroll the images.

The carousel is implemented in jquery, meaning it should be compatible with most modern browsers on the majority of platforms.

Design D25a: I created a logo for the banner:

The graphic is a binary sequence picture with the natural numbers on the bottom aligned against 'Number of groups of order n' - (A000001) on the top. More about sequence pictures here: http://www.mathpuzzle.com/MAA/07-Sequence%20Pictures/mathgames_12_08_03.html

I thought 'Sloane's' would provide an attribution for Neil and also address the informal usage of Sloane's to refer to the encyclopedia.

Design D25b: Some variants on the logo:

All type and logos are available as svg.

Comments

(from Martin Y. Champel) You proposed a lot of them, I like the first one best, there is lot to see and elegant. The idea of counting is well shown and clear even for non specialists.


This is awsome. Especially, that you took the effort to make a live version.

I am not sure if the direct link to the sequence is always a good idea. Maybe the link could go to a Wiki page with an explanation of the image, possibly showing a more detailed version. The sequences could be linked from there, and the Wiki page could also be linked from the sequence.

Sometimes there may be more than one sequence related to an image. For example I have shown the permutohedron as an example for A000142(4)=24. But the triangle row A019538(4,1..4) is an even more appropriate candidate for the link.

I really like your "record company from the 50s" logo (bottom row), but I think by no means it should contain a sequence, when there is a slider next to it. I would prefer a full width combination of name and logo, and a full width slider below.

Tilman Piesk 19:42, 25 January 2014 (UTC)

  • A lot of really great design work here. You illustrate the importance of considering the grayscale and black/white versions of any logo design. The carousel design is wonderful, and JQuery implementation degrades gracefully (I tried it in a 6-year-old Firefox, and the result is quite good). The images should be scaled down to actual presentation size (e.g. Sierpinski should be 180x120 pixels, not 600x400) and stored as GIF or PNG, because this stuff will be loaded millions of times per day. Robert Munafo 06:39, 26 January 2014 (UTC)

Design D26: Submission from Anne Ardichvili-Champel, 25 January 2014


It is probably not serious enough but still I see a link with D18 refering to elementary counting (stones or animals).

The child type of drawing also reminds us that counting is one of the first thing we learn. I guess all of us were told as kids to go to bed and count sheep to fall asleep.

The reference to the role of Neil J.A. Sloane emphasizes the fact that he not only founded it 50 years ago but still takes great care of the community.

  • I like your design. It is humorous and has an emotional effect. It is a nice counterpoint to the technical side of the database. Peter Luschny 11:20, 1 February 2014 (UTC)

I am pleased to see someone feeling the emotional effect in front of the sheep herd and noticing the counterpoint I tried to raise. (A. A.-C.)

Design D27: Submission from Arjun Jain, Jan 26 2014


It can be seen that my design is unlike others and is, to me, quite refreshing.

The idea comes from an art nouveau piece by Margaret Mackintosh that I saw some years ago.

I have integrated the design with some sequences mentioned on the contest webpage and the OEIS poster (A031214, A000108, A000001), by strategically placing whites and reds among blacks.

The design as a whole invokes intuitive feelings surrounding integers.




Design D28: Submission from Iakovos Ouranos, Jan 27 2014

I wanted to create something minimal, elegant and modern. The banner was implemented in Processing (http://processing.org - an open source programming language) except from the logo, which was created in CorelDraw. Before I decide eventually the final version of the banner I created about 10 other variants. Some static samples of my efforts are sample_01, sample_02, sample_03, sample_04, sample_05, also the following:




All these versions have the same underlying concept. Background rectangles have random positions, are distributed uniformly, though some of their motion parameters' values are the same. Local patterns are emerged by the combination of their motion and position. In a sense, these patterns correspond to the patterns implied by sequences in the realm of Integers. I chose to design a banner that has an abstract link to the integer sequences, a concept that somehow captures the big picture, the beauty, the harmony and the charm of numbers.

The logo is a vector graphic and I can provide various sizes or colours (a png file is available here). The main body of the banner is in code and thus various modifications in colours, sizes, positions of its elements and dimensions/filesize of the final file can be made with accuracy. I can also provide a Javascript version of the banner, in which the motion of the background rectangles is random and smooth.

I think that an animated gif would be very nice. The file size is quite large (~1Mb), but I suppose that the final dimensions of the banner will be smaller and accordingly the file size will be smaller as well.

Design D29: Submission from Pierre CAMI, Jan 28 2014

  • The picture is made using the number of way 2n can be expressed as the sum of two different primes, for the first 250000 values of n, ploted against the Euler totient of n.
  • Then using the colors of the rainbow for groups of 25000 n values: Violet for n from 1 to 25000, ..............., Red for n from 225000 to 250000
  • This is another way to see the Goldbach Comet.
  • If n is prime then the points are at the bottom of the picture,
  • At the top are the n values 3*5*7*11*..... = primorial/2.

Design D30: Submission from Pierre Lairez, Jan 29 2014

  • In context demo : index, A00108
  • The font face is Adobe Source Sans Pro Light, an open source font.
  • It would be very easy to reproduce this logo in pure HTML and CSS
  • This features a Latin square, an old combinatorial object.
  • The square is reduced : the first column equals the first row (however since there is colors instead of numbers, it is hard to say if the first row is increasing...)
  • This is anecdotal, but you will notice that the number of reduced Latin squares is A000315, which you can read A00OEIS, if you speak l33t.
  • I like the way that you've shown a mathematical object without using digits or graphs. It's somewhat subdued and very attractive. (Also, kudos for using an open-source font... kinda goes with our mission.) - Charles R Greathouse IV 00:12, 30 January 2014 (UTC)

Design D31: A submission from Gabriel Stadtler, Jan 31, 2014

OPTION 1:

OPTION 2:

Comments

  • The numeric acronym of OEIS forms an integer sequence.
    I think it would be better without the "," after the numbers, which separates the initials too much from the "rest". (IMHO) The coloring scheme makes sufficiently clear that there is a sequence of numbers to be read vertically. — M. F. Hasler 02:45, 3 February 2014 (UTC)
    Thank you for the observation. I included it as another option. — G. Stadtler February 11, 2014

Design D32: A submission from Robert H. Barbour, Feb 1, 2014


COMMENTS

  • The OEIS has many of the attributes of Galaxies like our Milky Way and other Spiral Galaxies. The OEIS is a Galaxy of number relationships occupying `number space'.
  • We have a solid grasp of some relationships as detailed in OEIS but vast areas are a blank canvas much like the the dark areas in a Galaxy. For example, primes are common in the low integers and in the same way as spatial objects are common at the centre of Galaxies but rare in higher integers as are spatial objects away from the centre of Galaxies.
  • Number relationships seem to come in families much the same as the spatial objects in the arms of Spiral Galaxies.
  • A picture of a Spiral Galaxy is easily obtainable but it may be most appropriate to create a cartoon effect/ model rather than a photo.
  • I liked the Times Roman font for the text in M F Hasler's logo D14 series so have used it in my suggestion, along with the wording which is succinct and gives credit where due.


Design D33: Submission from Andreas Boe, Feb 2, 2014

Design D34: Submission from Mihir Mathur, 2 Feb 2014

The logo was reworked by Abhinav Thukral. It is a form of a box, which is emitting sequences.

Sample:

Design D35a, D35b, D35c: Submission from Russ Cox, Feb 2, 2014

I suggest that we standardize on the basic headline and that we include the search box in that headline, which will save space on the rest of the page. Then we can use the space to the left and right of the headline to showcase two sequences. The artwork should be just black and white (faded to gray by the system) to avoid detracting from the actual page content, but it should be interesting and if you click on the artwork you should get to the information about the sequence. Each day the two sequences might change. If a sequence does not visualize particularly well it would be fine to list some terms from the sequence instead. The image shows the effect for three different choices of sequences. The three different banners also experiment with the subheading.

File:RussCoxBanner.png

(Discussion) Some remarks on the layout from Peter Luschny, Feb 01 2014.

An assessment of the banner is really possible only in the context. That's why I built an Html page that shows the banner on a sequence page. In doing so I noticed that some things of the current layout are in need of revision. The most important are:

  • To plunge a navigation bar into the data row is one of the most absurd designs throughout the Internet. It deserves the Golden Lemon Award by the W3C. Please give the links "list graph refs ..." a line on its own!
  • The page footer is completely crammed. There should be no more than two lines of links. Every other page might be reached through the link 'other pages'.
    I agree. Further improvements(?): (i) The "Register" link (used once in a lifetime) could be RIFO "Login/Register" which would go to the login page where "Need an account? visit..." should be rephrased ["(If you are) Not yet registered? Please follow this link to request an account."]). (ii) Instead of the "Lookup" link, better put a search bar on each page, as on the wiki. Anyway, a click on the OEIS logo takes to the homepage = lookup page. (add title="Home & lookup page"). — M. F. Hasler 02:41, 3 February 2014 (UTC)
  • I added a link to the 'editors' in order to make them better visible and changed the label of the link 'more' to 'help extend' and expanded 'seq.' to 'sequence'. [Link added by MFH, cf below.]
  • I think it increases the readability and the navigation to put all links in a smaller font and in one color. These red links look as if they were designed in the style of the 90's and we can be glad that they don't even blink. In fact these screaming links interfere only when reading.
  • I implemented the demonstration page in XHtml 1.1. According to the [W3C-validator] it is error-free. The OEIS page on the other hand is implemented in HTML 3.4 and has 21 [errors] according to the validator. I think an organization like OEIS should deliver error-free Html.
  • Here is the [fake] (pretty nice! let's call it Design D36. - Neil Sloane) and here is the [original] so you can look at the pages side by side in your browser. Compare with my favorite [competitor].
    That info ("fake link") took me quite some time to find! should be linked to already earlier: I permit myself to make that edit as a suggestion. — M. F. Hasler 02:41, 3 February 2014 (UTC)


Design D36 from Francesca Librandi, Feb 12 2014

Design D37 from Andreas F. Raseghi, Feb 17 2014

Comments

  • My design goals were to keep the site header classic and relaxed (not overloaded) and to use colors, that encourage users to try a sequence.
  • The OEIS name logo was constructed with a dotted number "1" instead of a letter "I". Because unity is the basis for integers, this emphasizes the mathematical nature of the site.
  • The "fast forward" symbol stands for both fun (of searching) and speed (of finding). And in fact, the OEIS site is very effective! :)
Personal tools