Smythe:
Here’s one attempt:
http://www.answers.com/topic/laoco-n
and here’s the picture:
http://en.wikipedia.org/wiki/Laocoön
Another name I’ve always found hard to pronounce is ‘Pirithoo’ (ablative of Pirithous I think) which is the last word in Horace’s spring Ode 4.7:
http://www.merriampark.com/horcarm47.htm
Wish someone would decline Pirithous for me! 
Cheers,
Int
In case you’re not joking, it’s second declension: Pīrithoüs -i (Pī-rí-tho-us, 4 syllables). L&S & OLD.
Ut non joceris, secundae declinationis est (Pī-rí-tho-us sonitur, nisi fallor, quattuor syllabas habet, secundum L&S et OLD).
I haven’t seen any numbers, but from having studies the lists of Bennett and Buck from almost a century ago, and compared them with more recent publications like Allen, I would guess that there might be a at most a couple of dozen words that are at least moderately common, where there is a vowel whose hidden quantity is in dispute. I don’t think it is a large problem; an ambitious editor (cough) might want to investigate each of them and compare different arguments; or you can choose to follow a particular dictionary and stick to it; or investigate what a majority of publishers do and follow that; or simply leave them unmarked. As much as I am interested in historical authenticity, I don’t think it is a very important issue.
Adrianus:
I wasn’t joking 
so thanks. I suppose I deceived myself into thinking that because it was a Greek import it couldn’t be that simple. But then, even Lewis & Short online seem a bit confused when it comes to the masculine form:
pirithoum noun pl masc gen no user votes 41.1% [vote]
pirithoum † noun sg masc acc no user votes 58.9% [vote]
† This form has been selected using statistical methods as the most likely one in this context. It may or may not be the correct form. (More info)
Anyway, now I can enjoy the poem all the way to the last syllable.
Cheers,
Int
Greek import words are tricky because sometimes there is more than one attested form. For a trivial example, athlētēs (a first-declension noun) was sometimes declined regularly as athlēta – maybe much more often, in fact, since Cassell’s Latin Dictionary has only athlēta.
I’ll say something here in a few days.
My name is Michael Everson; I am a font designer and character encoder who has a certain expertise in linguistics, and I am the publisher of the book that is being discussed in this thread. Shortly I will post a link to a PDF of a sample of the new publication.
And the book is…
Alicia in Terra Mirabili translated by Clive Harcourt Carruthers!
FIrst published in 1964, this splendid book is very hard to get. I publish a range of Alice materials (translations currently available in Cornish, Esperanto, French, German, Irish, and Manx, with Italian, Swedish, Scots, and Ulster Scots in the works.
Alatius and I have been discussing the presentation of this publication, which inspired his poll. The results of that poll were very interesting—not least for showing some differences in taste between Anglophones and learners of Latin who speak other languages.
Please see this sample for a comparison of the text with macrons/j/v and with i/v.
Very cool. Looks good! When shall it be available?
Well, it will take a bit of time to put together. Perhaps in time for Christmas.
Yaay! Christmas present for myself. I can’t wait.
I just found the survey. And I just recently discovered after some thinking that macrons are really really important. Before I just threw them aside as unimportant – but they kinda tell you everything. I’ve taken Japanese as well, which also distinguishes between vowel length, and I would think it to be just silly if they didn’t add in the extra us in places. So yes, I’m glad the survey goers had the same thinking as well. ^^
Has anyone else had a chance to look at the sample?
If you created the fonts, they’re nice. You’re the publisher so what’s the copyright like in this translation? I see it here http://www.intratext.com/IXT/LAT0697/_INDEX.HTM on the internet, of course, with a 2007 copyright that can reside only in the concordances, but the translation copyright must be 1964.
Si typos creavisti, macte. Es editor huius operis. Quid tunc de jure interpretis? Cunctus textus hîc interrete—http://www.intratext.com/IXT/LAT0697/_INDEX.HTM—reperitur ubi annus bis millesimus septimus legitur (quamvis solis prae concordantiis ille annus jurem attendat) at jus interpretis anno millesimo nongentesimo sexagesimo quarto detur. [/i
I personally prefer the first version, with the macrons and j’s. But that’s because I’m at such a low level of proficiency in Latin, so I like all the help I can get, and macrons and distinguishing i’s and j’s both give me help.
They both look very nice, though. And I’m glad to see you’ve kept the illustrations.
Correct me if I’m wrong – which is very possible, but this rule will work at least 99% of the time – but I think “i” is always a vowel except when it occurs as the first letter of a word (where it is always a consonant if it precedes another vowel), or when it occurs right after a prefix (example: adiuvare, from ad + iuvare – prefixes should be easy to recognize since most are also prepositions). So “iam” starts with a consonant since it’s the first letter and followed by a vowel, but it’s a vowel in “diū” because it’s not the first letter and “d” isn’t a prefix. It’s a pretty easy rule to handle, I think.
If for any reason you’re unsure whether an “i” is consonantal or not, you can look it up in Whitaker’s Words. For example, if you type in “iam”, it will return “jam”.
Well, usually, but there are exceptions: for example, the son of Aeneas is “Ĭūlus” (three syllables). In particular, words of Greek origin do not have “j”.
I added the macroned glyphs to the fonts, but did not create the fonts themselves.
You’re the publisher so what’s the copyright like in this translation?
The copyright rests with the Estate of Clive Harcourt Carruthers. I met the Estate (Clive’s son and daughter-in-law) on 21 May when they passed through Westport on a bus tour through Ireland. I have permission to re-publish both Alicia in Terra Mirabili and Aliciae per Speculum Transitus.
I see it here > http://www.intratext.com/IXT/LAT0697/_INDEX.HTM > on the internet, of course, with a 2007 copyright that can reside > only > in the concordances, but the translation copyright must be 1964.
I’m afraid that whoever has put that text on the internet is in breach of the Carruthers’ copyright. Not only is the text incomplete (there are twelve chapters) but it is uncorrected. And unauthorized. That’s not right.
Carruthers lived from 1891 to 1980, so the copyright rests with the Estate until 2055.
“Quid adjuvat liber,” sēcum reputābat Alicia, “in quō sunt nūllae tabulae aut sermōnēs?”
Yeah, yeah, yeah… I still like the reinforcement of having i and j distinguished in the text. Much easier than interrupting oneself by looking up the word in Whitaker’s or whatever else. As I said earlier, I’d even take accent marks in Latin, if they could be had. Anything to make the language easier to learn, I am for.
And, to be honest, I say if you’re afraid of learning very simple rules like “‘i’ is a consonant 99% of the time when it’s at the start of a word or after a prefix and followed by a vowel, and always a vowel otherwise”, then Latin isn’t the language for you. 
Looking stuff up in Whitaker shouldn’t even be necessary because the rule is so simple.
I mean, Latin is full of rules you have to memorize. Where stress goes (when you already know which vowels are long) or when “i” is a consonant are tiny pebbles atop a mountain. Taking off a couple of tiny pebbles for one person at the expense of making everything look funny to everyone else doesn’t sound like a good compromise to me.
To be fair, I can certainly see the rationale for “j” and I think it’s actually more logical than using “i”. “Different phonemes should have different graphemes” is a good convention to follow when designing an orthography. Nonetheless, whether a text uses “j” or not is a pretty small matter, certainly not on the scale of whether it should use macrons (since whether a vowel is long cannot be inferred from context, unless you’re reading poetry and can figure out how the line should be scanned) – hence I think it would be wise to defer to common convention, and few write Latin with “j” anymore.