Wiktionary talk:Main Page/2009 redesign/Part 2

From Wiktionary, the free dictionary
Jump to navigation Jump to search

On icons, is there any way we can stay away from the books? I'm tired of those icons as a book doesn't convey any information. Also there are several sections where a globe would apply and I think we need to do better than saying this is where the globe goes. That one could really be tailored as with WT:TOW. We've had a few interesting ideas but it would be nice to have more. I suspect there would be many others if solicited. Can we do a side-by-side comparison? DAVilla 18:17, 29 April 2009 (UTC)Reply

Could we have an easy-to-find link to simple wiktionary ? There are anumber of new users who would probably benefit from this. -- ALGRIF talk 10:20, 30 April 2009 (UTC)Reply
I was wondering about OmegaWiki as well. There doesn't seem to be a link to that anywhere. 72.177.113.91 13:53, 30 April 2009 (UTC)Reply
[edit]

I like the look and the ideas. Quick impressions:

  1. Don't users expect a sort of ToC for the page itself?
  2. The icons seem a little big and cartoony.
  3. There might be a little too much whitespace on the above-the-fold screen.

OTOH, it's really quite attractive and leads new users to where they can get the most value. Is there any way to test it on a "small" population? DCDuring TALK 01:56, 29 April 2009 (UTC)Reply

  1. We would be the first Wikimedia site (heck, one of the few MediaWiki wikis around!) bothering with having one...
  2. The icons are the same size as the original ones. They are only more numerous above the fold (5 vs. 4). removing some leads to either unattractively unaligned titles, or unattractive whitespace. These are a personal choice, feel free to come up with suggestions (besides, they can probably be significantly de-cartoonified when they are redrawn to unified hues)
  3. It's an illusion due to the fact the big, useless wall of text has been removed and the index on top pared down. There is much more actual explicited content. Also, note that everything will move upward significantly on the actual page, since I cannot remove the page title and the subpage links.
I have NO idea how to "test" it (besides asking the usability workgroup thingie at the foundation to look at it), though I'd love to get more comments from average users of the site, as opposed to very involved editors. Circeus 05:30, 29 April 2009 (UTC)Reply
  1. It would be nice to show a little leadership on basic usability on the main page, if not in the entries themselves.
  2. and
  3. I wasn't comparing to our current page. Do the layout and icon size contribute to the white space?
It might be worth asking them before tweaking further. The more effort put in, the harder to remain open to new points of view. I would be interested in hearing about your experiences with them. DCDuring TALK 10:10, 29 April 2009 (UTC)Reply

Funny, I think there's very little whitespace. The only significant bit is below the left column, where it is shorter than the right. Adding a bit might improve clarity.

Don't get to uptight about the fold. Haven't you heard, there is no fold, at least not since 1997Michael Z. 2009-04-29 16:30 z

My understanding: The "fold" only matters when users are "just testing" a site. Nielsen-Norman research has found (recently) that, in that state. users still don't page-down. Of course, my contract here provides or a big percentage increase in pay based on increased usage. You may only see the downside: patrolling, more amateurish entries, more admins, et al. Need I say more? DCDuring TALK 16:46, 29 April 2009 (UTC)Reply
Yeah, we were only using the term here to define the top portion of screen. I've never designed for the fold before, and I'm not going to start now. Circeus 19:59, 29 April 2009 (UTC)Reply

I feel like there's a bit too much whitespace in the right column, in comparison to the left. If you follow Word of the day across it's not even with More than words. Likewise there's a lot of emptiness in the nebulous area to the right of Participate. I'm not saying that white space is bad, only that it should be better distributed. DAVilla 18:33, 29 April 2009 (UTC)Reply

When the icons are standardized then the first heading alignment should be equalized too.
I agree that the whitespace at the paragraph level is pretty chaotic in the right column. Especially the double-width indents – in my browsers, the indent is about the size of the defined term “Rhymes”. The boldfaced terms define the paragraph hierarchy very well, and I don't think any indent at all is needed. I'll take a stab at this. Michael Z. 2009-04-29 19:01 z
Okay, now the text helps define the left edge of the right column. Not only was most of the text indented by a full word-length, but the top and bottom box had different left margins. Michael Z. 2009-04-29 19:51 z

Okay, where the frick did you get the idea that "top and bottom box had different left margins"?? The bottom small-sized texts did not have any margin, they were centered!

Also, trust me on that, the headers are all precisely flush to each others (unless that changed in your edits). The only reason they might LOOK not to be is that the first letters might not look like they "start" at the same place (put a W or a Y, and a O flush one above the other and you'll understand), and that the space between the icon and the text might be or look wider depending on the shape of the icon. I'm fairly sure the folder should be moved back a cuple pisxls to the left, though. Also I definitely don't like the resulting look on the right side. Circeus 19:59, 29 April 2009 (UTC)Reply

Have a look at the revision before I started. Notice how Wikisaurus, Appendices & glossaries, Rhymes are indented from the heading's underline, while Information desk, Tea Room are not? That's where the frick, friend. You didn't even notice it because more than half the lines had crazy 4-em left margins.
I fixed it in this edit.
I haven't made any changes to the headings or icons this week. You'll have to be more specific about “the look on the right side,” because I can't tell what that means. Michael Z. 2009-04-30 00:38 z
Em dashes are used for attributing quotations, but they are not bullets. The bold weight of the subheadings and line space between items serves to show the hierarchy here adequately. Adding em dashes serves no useful purpose, and the choice looks unprofessional. Michael Z. 2009-04-30 00:42 z
Can you take the em dashes, out, but leave in a bit of indentation? DAVilla 01:08, 30 April 2009 (UTC)Reply
Done, but I think it may be better with a full line space separating the paragraphs instead. Michael Z. 2009-04-30 01:37 z
I've just tried adding vertical space instead. The outdenting is non-standard and breaks up the left edge, because the boxes contain as many boldfaced subheading lines as non-bold descriptive sub-subheadings. The result is a chaotic left edge. Compare indented presentation to line-spacedMichael Z. 2009-05-02 14:34 z

Define button

[edit]

Can you add alt text to the Define button? How is it different than the Go and Search buttons on the left navigator? If it is the same as the Go button, will the different name be confusing to users? --Panda10 11:51, 29 April 2009 (UTC)Reply

I wasn't sure whether it wasn't meant to go to a new-entry creation screen. DCDuring TALK 14:00, 29 April 2009 (UTC)Reply
I don't really like that feature up top in the first place because the search box to the left does a much better job of it, including auto-completion. If it's there, it's probably better to be consistent. DAVilla 18:37, 29 April 2009 (UTC)Reply
Yeah, we shouldn't give the same interface two different names & labels, whether on the same page or elsewhere. An alternative would be to change (Go) (Search) to (Define) (Search), but then we would be inconsistent from other wikis. Really, the reader is going to an entry, which defines, and possibly etymologizes, synonymizes, translates, etc.
Define might be reserved to describe the action of some widget where just an extracted definition line is swapped into the page without a reload, or something. Michael Z. 2009-04-29 18:53 z

Trimming down the supplementary bits

[edit]

Any objections to trimming down some of the external-link bits?

en.Wikipedia's main page has foreign-language links to the largest 55 Wikipedias with 20,000 articles or more. We have bilingual links to 107 Wiktionaries, with 100 entries or more (heck, even a monolingual pocket dictionary has 40 k to 150 k terms). Our list takes nearly twice as much vertical space as Wikipedia's, and almost a full window-height in my browser. We both already link to Meta's up-to-date full listing. I think we only need to give foreign-language readers a taste and a pointer, not try to duplicate meta's comprehensive coverage right on our home page. Remember that each of our entries also includes FL links. May we reduce our front-page links to FL dictionaries containing at least 10,000 entries? Remove the translations?

commons:Main Page has a rather attractive array of sister projects at the bottom, which sports the same information as ours, but in a more compact form with small icons. Can we adopt this format? Michael Z. 2009-04-30 15:35 z

There's a pretty good case to remove at least the ones under a thousand. It's been years already, and if they haven't got there yet, there's just no way they can keep up. We do want a link to Simple English though, regardless of the numbers.
Personally I like the big icons better. 63.95.64.254 03:25, 1 May 2009 (UTC)Reply
Three points to consider: (1) Wikipedia is a topical project, where articles are all in a single language. Any information in another language's project isn't particularly relevant most of the time. Wiktionary has the goal of all words in all languages, so the content on those other projects is immediately relevant to us and our users. (2) Meta's list is not all that up to date most of the time. There is an independently generated page that is more up to date than Meta's own page. (3) We had 10 new FL Wiktionaries break 100 in the last update that I did. These smaller, fledgling Wiktionaries start well, but without visibility they tend to stall. I personally like seeing a link to them from our page, since it can help to stimulate interest in them. --EncycloPetey 03:38, 1 May 2009 (UTC)Reply
I don't understand your point 1. Just like Wikipedia, en.Wiktionary is written in a single language. Entries in other-language Wiktionaries are just as relevant to us as other-language Wikipedia articles are there – if they have information we don't have, then we may want to transfer it here for our readers. I don't read Georgian, so the Georgian dictionary can't tell me anything about either Georgian, English, or Pashto words, no matter how relevant the information is to me.
If I'm not able to decode “ქართული”, then there's no point in adding the translation “(Georgian)” on the home page for me. Just like we don't translate it in the “in other languages” navigation in the sidebar of every entry.
Re 2, then let's update Meta. If we can't and the meta list is unsatisfactory, then let's link to the external resource instead. We don't have to be greedy about linking to make money, so let's best serve the reader.
Re 3, if a purpose of the list is to promote Wiktionaries with potential, then why not add the full list of 172? Conciseness is good, but if this is really useful, then we can be comprehensive and promote the ones where a contributor can make the biggest proportional impact. It's not like scroll bars are a scarce resource. Michael Z. 2009-05-01 14:09 z
(1) Not true. If there are English translations or example sentences, I can make use of FL Wiktionaries in languages I do not speak. This isn't true for titles of Wikipedia articles, where we can't be sure of the meaning of the title even if it links to a particular English Wikipedia article.
(2) Meta is always behind the external source I mentioned, and has been for years. They simply do not update often enough to keep their list current. Linking out to an external source will not assist all the various other Wiktionaries, who usually rely on our list to keep theirs up to date.
(3) Because there does need to be a cutoff, and some projects have 0 entries. I think the appropriate cutoff is 100, just as it has been. --EncycloPetey 03:52, 2 May 2009 (UTC)Reply
(1) I just don't get it. Yeah, I guess someone could use the English Wiktionary home page as an index for randomly finding citations in other Wiktionaries. Does anyone do this? Does having this bilingual list on our home page demonstrably serve this purpose for even one reader?
I still see no reason for including the English translations. Let's cut it in half by removing them, or moving them to tooltip title attributes. Michael Z. 2009-05-02 14:26 z
Many users do not know the native names of languages, and even I don't recognize a lot of them despite having worked with that page more than most people here. The English names allow users to see which Wiktionaries are listed and where. --EncycloPetey 20:36, 3 May 2009 (UTC)Reply
Ah. So we have a lot of English-only readers who come to our home page looking for the Polish-language dictionary. We can serve them better by translating Polski on the home page. This fulfils what, number 900 on our list of priority home page objectives? Boy, there's nothing as pleasing to the eye as a good focussed design. Michael Z. 2009-05-04 03:41 z
So your primary objective is to be negative? I don't see any harm in the listing, and it's useful. It's already up there, so placing it at number 900 on the to-do list is irrelevant. The work is already done. The page will still be long even if we remove that information because of the interwiki links down the left side. On all the monitors I've used, that side list runs past even the current content. We might as well use the space with the content we have rather tha present a big blank space with no information. --EncycloPetey 00:59, 8 May 2009 (UTC)Reply
Gee, there's still some blank space there, even with the English translations for foreign-language readers. Why don't we throw in some metric conversion tables and the numbers of our local pizza joints? Not only is there no harm in that, it would actually have helped me today.
No, my objective is to improve the design. You don't make something excellent by adding stuff just because “there's no harm in it.” “We might as well” is not a confidence-building design strategy. Michael Z. 2009-05-08 06:49 z
You're still proposing a change based on a subjective criterion. I believe that removing the links will detract from the Main Page, you believe retaining them will do so. It's fully opinion and nothing else, and what's more, only the two of us seem to care either way. --EncycloPetey 03:08, 9 May 2009 (UTC)Reply
I guess you're right on the last point, but that just means we have the luxury of only having to convince one other editor.
Subjective? I suppose so. But based on the objectives of the main page, basic design principles, and simple logic, I would still say that I am right. I guess there's no point in arguing design at all, because one editor just has to demand that another test 41 shades of blue to stall it. And our home page will continue to look like a design by a committee. Michael Z. 2009-05-11 07:22 z

Two issues

[edit]
  1. Should we add RfV to the "Participate" section? That really is one area where even a complete newcomer can help, and it is often underused.
  2. The icon for the "Word du Jour" looks like cursing. Strings of punctuation characters such as !@#$ usually indicate swearing, and that conveys the wrong impression for that section. --EncycloPetey 20:34, 3 May 2009 (UTC)Reply
Heh, so it does. Perhaps, instead of strings of symbols, we could put a couple of simple, stylised images in the speech bubbles, representing the kinds of basic but useful term we might include there? A house etc. Equinox 10:17, 8 May 2009 (UTC)Reply
That was the best thing I could find. It's not really a final choice. Circeus 16:15, 8 May 2009 (UTC)Reply
Or empty balloons, letting the reader “fill in the blanks.” The icon is a bit busy too – a single balloon would get the idea across, and the oval background is pointless unless it is used to graphically unify every icon. Michael Z. 2009-05-08 16:51 z