Massive error in Wikipedia

charles ellertson's picture

I put this in the build section because the people with the most experience tend to hang out here, at least from time to time.

* * *

If you look up most any typographic topic in Wikipedia -- say, letterspacing

http://en.wikipedia.org/wiki/Tracking_%28typography%29

or

justification (typographic)

http://en.wikipedia.org/wiki/Justification_%28typesetting%29

you find so many errors as to want to cry, shout, whatever your favorite form of expressing frustration.

It's beyond my energy, and competence, to fix everything, but we should try.

Nick Shinn? John Hudson? Etc.? Others who who know the history & how technical word usage has changed over time.

For example, the claim that:

Kerning adjusts the letters closer together (negative spacing), tracking adjusts the letters further apart (positive spacing).

ignores the history & development of the use of this technical term. Originally, with both foundry type and Monotype, kerning was always a removal of space between letterpairs. -- a saw with foundry type, and casting a kern with Monotype (& probably could have foundry-cast type with a kern, don't know if this was ever done).

The point is that the original notion of kerning always being a negative space was simply a technical issue, hunks of metal were involved. You could add space, but that was a different technique.Can't do it with a saw. And in no case (save perhaps Monotype?) would you apply positive space to a whole paragraph, just too damn much work.

As photocomp made inroads, letters were no longer hunks of metal, and the same system that provided for removing space between pairs of letters could be used to add space. "Postive kerning" ceased to be an oxymoron.

And on & on.

If Wikipedia is to be fixed, it's going to take a bunch of us. Not interjecting our personal aesthetic, just accurate, fairly complete description.

In the "talk" page on justification, Wiklipedia says

This article has been rated as High-importance on the importance scale.

and my first comment is

This article is full of inaccuracies

Anyone want to help?

hrant's picture

Actually a "kern" originally meant part of the face of a metal sort that extends beyond the edge, like the beak of an "f"; by extension "kerning" could mean filing down the side of a letter to allow a tighter fit (while allowing/increasing a kern).

hhp

Nick Shinn's picture

I’m afraid I don’t consider that my responsibility, just because I talk shit at Typophile.
Besides, none of my typefaces have Wiki pages.
Ask David Berlow or H&FJ, theirs do.

charles ellertson's picture

I’m afraid I don’t consider that my responsibility,

Oh, well, I am much relieved. I didn't realize it was only a matter of responsibility...

Rob O. Font's picture

I have been there, and done that. The massive problem with a democratically elected definition is that it's like a bad dog... it just doesn't "stay". Even discussing this issue opens my past wiki work up to entities who just want to frustrate the public.

When I found this issue, and realized who was involved, I tried to whip up enthusiasm for the wiki here, but that's failed. Good luck.

hrant's picture

David, you're certainly right that Typophile is the best place to [try to] do it right.

So: I'm game to give it a shot (which I've never really done before).

Can we get a volunteer show of hands?

hhp

Karl Stange's picture

I am not the best person to provide typographic or technical information but would be happy to lend administrative support. I have learned so much from this forum (specifically the good will and hard work of contributors) but the current wiki functionality/layout is patchy at best.

charles ellertson's picture

I'm sorta, kinda willing. These days, if I have to think, my head hurts. If I have to write down what I think, it hurts a lot. Always painful to (try to) communicate. And I just finished writing a chapter for Hendel's book, that's my quota of botched communication for the year.

OK, enough whining. I'll try to help.

David's comment reminds me of a cartoon (in the New Yorker, I believe).

Two dogs are cavorting in the park. The first turns to the second and says "Hi. My name's NoNoBadDog. What's yours?"

oldnick's picture

I think we should let coachbag handle this problem…

aluminum's picture

One way to make things a bit 'stickier' on Wikipedia is to use citations. It might be more work but updating the wiki here and then using that as a citation on Wikipedia might be helpful.

HVB's picture

That's circular validation - "I said it here and then referenced it, so it must be correct".

Chris Dean's picture

Here is an example from a primary source with a proper APA citation:

————
Justify: To adjust the length of the line so that it is flush left and right on the measure. Type in latin alphabets is commonly set either justified or FL/RR (flush left, ragged right).

Bringhurst, R. (1996). The Elements of Typographic Style. Second Edition. Point Roberts,WA: Hartley & Marks.
————

Do this 10 times for each term, each from a different source, and Wikipedia may consider updating their content.

Thomas Phinney's picture

Wikipedia is not a “they” and one does not need 10 citations to change something. If there is content that is wrong, go change it! Anybody can. If the incorrect content has a citation, try adding two for your version. If not, add a citation if you've got one handy.

If somebody is arguing and reverting your corrections, then there is something worth complaining about. But don’t make it sound so hopeless. It is pretty easy to fix most things on Wikipedia. Usually, only when something is controversial (or possibly incorrectly perceived as such) is it tricky.

Thomas Phinney's picture

Further to this, none of the information in the existing Wikipedia article had backing citations anyway. So I just spent less than 20 minute and largely fixed the darn thing. Feel free to go and edit and do further clean-up if you like.

Thomas Phinney's picture

Here is how you can compare my edits vs the previous version:

http://en.wikipedia.org/w/index.php?title=Letter-spacing&diff=537587559&...

HVB's picture

TP: - Nicely done! From a position of ignorance, I wondered why you said that metal type could only be negatively kerned. Wasn't it possible to spread two letters apart? In fact, unless the individual characters were physically altered (by removing parts) they could NOT be kerned by moving them closer together. Granted, that was the only meaning of the term at the time. I'd suggest saying it something like this:

"In its original meaning as defined for metal type, a kern could only bring letters closer together (negative spacing my modifying characters such as A and V so they weren't rectangular), though today it can go in either direction. Tracking can similarly go in either direction, though with metal type one could only adjust the letters further apart (positive spacing)."

I wouldn't deign to make such a change myself, because I really don't know whether I'm right or not.

- Herb

hrant's picture

Please read my first post concerning what "kern" means in metal type.

hhp

Thomas Phinney's picture

What Hrant said. Although it was certainly possible to space type. One could tweak it to say that what we today would call positive kerning was possible in metal type as well—it just wouldn't have been called kerning back then....

I will go back and tweak.

Nick Shinn's picture

…though with metal type one could only adjust the letters further apart…

IIRC, the entire text of The Book of Oz Cooper was “negatively tracked” by shaving each piece of type.

hrant's picture

In the 60 point Pascal that I bought some years ago a large proportion of the caps are filed down (some of them in a staggered fashion). I remember Gerald Lange (my letterpress teacher) saying that you can tell how accomplished a letterpress printer is by how much filing he's done!

hhp

Chris Dean's picture

@Thomas Phinney: When did you make your edits?

Thomas Phinney's picture

Right when I said I did. Why? Wikipedia's edit tracking tells you who did the edits and when the versions were saved, and shows you what was changed. You don't need to know the exact time to do this....

charles ellertson's picture

Thomas, one small quibble, and it's probably not for Wikipedia.

You corrected the Wikipedia copy to Although digital type sets tighter on average than metal type...

All your changes were for the better, I think, but I'd at least qualify this one. The overall wordspacing with computer comp (esp. PostScript > OpenType) is probably tighter then with metal, but the letterfit with metal was often tighter. You are of course quite right that letterfit is a function of type design.

With both Monotype and foundry metal, the presence of kerns on some letters meant that type would touch, though not overlap. And that meant editors -- those who mark proofs -- when looking at metal, were comfortable with "touching" letters. When photocomp came along, "not touching" became the new standard for text settings. For fun, just look at some Penguin books during Tschichold's and Schmoller's tenure.

As for Linotype -- Over the past week, I've been re-reading a couple Nero Wolfe novels.

The first, The Mother Hunt, (1963) was obviously printed letterpress, and just as obvious, printed from stereo plates. Too many copies printed for it to have been printed from the cast type.

The second book, The Doorbell Rang, (1965) is set in the same typeface. My copy is a first printing, first edition. While it is set from the same metal type, I'm 99.44% sure repro was pulled and the book printed offset. It has that look, and an (almost) 21.66 pica measure offers another clue.

Here's the point. These were set Linotype, not Monotype. Ignoring the problems of certain characters with the linecaster, the letterfit is tighter than we usually see today, esp. with the book where repro was pulled, where the ink spread is a bit greater.

The common "not touching" requirement that came in with photocomp, even with kerning regularly applied, often results in a bit looser letterfit.

Thomas Phinney's picture

Hi Charles,

I'm sure it could be wordsmithed and further improved: feel free to do so! It’s Wikipedia, that is how it is supposed to work. I was trying to balance being succinct and being accurate, but I think you are correct that it can be made better. Please do! :)

charles ellertson's picture

Thomas, I think what you've done is great. The comments I added here are for Typophiles, not a general audience. Too much reliance on, what, "unspoken industry practices"?

Maybe it could be wordsmithed better, but as I so often give evidence, not by me. However, since you've done some good work, I'll take the inspiration to heart & try to improve anything in Wikipedia I can (maybe) make clearer.

Thomas Phinney's picture

Yay!

Wikipedia is far from perfect. But in general, in most articles/areas, it tends to improve over time.

hrant's picture

Indeed.
And as if just because something is printed on dead trees it must be reliable.

hhp

charles ellertson's picture

OK, Thomas. I've just begun on "Justification."

http://en.wikipedia.org/wiki/Justification_%28typesetting%29

Only down to

"People with dyslexia..."

Going to have to look up the Adobe patent on paragraph optimization, because the routines of TeX are attributed to InDesign, which is sadly wrong...

Anyone else going to help? The whole damn thing is from the point of view of a LaTex user who is an author, not a typesetter, type designer, etc.

& I don't know any system or fonts that use Zapf's alternate glyphs, anyone?

charles ellertson's picture

spam filter/double-post ...

Chris Dean's picture

Has anyone thought to, while developing content for wikipedia, repurposing the same content for Typophile’s Wiki? Two birds, one stone…

Nick Shinn's picture

There is also a PrintWiki.

Syndicate content Syndicate content