Shavian eGroup Archive Browser

----- Original Message -----
From: Scott Harrison <scott_harrison@...>
To: <shavian@eGroups.com>
Sent: 16 December 1999 8:29 pm
Subject: [shavian] Re: Homographs

> I have written an automatic translator that I am using to generate
> Unicode files of Shavian data. The first two translated documents are
> already on my web site.

Where's that?

> In working on this I have discovered there are
> many words that prove troublesome (e.g., record(s), object(s), produce,
> et al.). The way I handle this currently is to leave them in Latin
> letters and then read the appropriate passage to determine the context
> and therefore which Shavian to use. Note that this is currently a
> manual process but as I get into larger documents I will codify what I
> do to attempt to get it more automated. But who has the time...

Almost all of the homographs I have come across so far differ in respect of pronunciation and grammatical role (e.g., live: /ai/ vs. /i/, adjective vs. verb). In order to automate the transliteration completely and to codify what you do when you check the text manually, surely you would have to parse the entire text, determining the grammatical role of each word? That's certainly more than I want to bother with in my program. I said "almost" before, but sometimes the pair of homographs have the same grammatical role: e.g., "lead" (n.), a leash, the thing attached to a dog's collar, and "lead" (n.), the metallic element or the graphite stuff in a pencil. That would make things even more difficult.

Since I wrote my previous message, I've added a third optional field (for alternative pronunciations) to my database. Now, whenever my program comes across a word-form that has more than one possible pronunciation, it gives the user 50 characters of context and asks for a decision as to which pronunciation to use. It does all this very politely and efficiently, and might even play Muzak in the finished version. In practice, this interruption doesn't happen very often, and serves to wake me up every now and then during the longish process of transliterating a chunky text.

How does your automatic translator work? Is it based on the CMU pronouncing dictionary?

BY THE WAY, DOES ANYONE KNOW IF A BRITISH RP PRONUNCIATION DICTIONARY EXISTS IN COMPUTER-FILE FORM???

Lionel Ghoti

------------------------------------------------------------------------
Was the salesman clueless? Productopia has the answers.
http://clickhere.egroups.com/click/1702



-- Create a poll/survey for your group!
-- http://www.egroups.com/vote?listname=shavian&m=1

Hello,

I have noticed people posting Shavian information using ASCII characters, and therefore would assume people view this information by switching the default font to one that renders those characters as the appropriate Shavian glyphs.

However, I have started putting Shavian data onto my web site as Unicode files that use the current ConScript encodings (as the final code points for Shavian are not defined). Are there people out there that can read the Shavian Unicode information? If not, do you know ways that one can display Unicode files on your platforms in general? I can always attempt to help figure out ways to look at the data in Unicode format since that seems to be the best longterm solution.

In fact, one should be able to create HTML pages that are in UTF-8 with mixed Shavian and any other alphabet on them. If so, we just need to determine how the browsers can be convinced to use specific Shavian fonts when encountering Shavian characters on a web page. Again, I am willing to help people be able to do this.

--Scott

------------------------------------------------------------------------
Get 75% Off Our Best Selling Health and Wellness Books!
SelfCare.com has everything you need to take care of you and your family,
plus spend $40 or more and your shipping is free!
http://clickhere.egroups.com/click/1824



eGroups.com Home: http://www.egroups.com/group/shavian/
http://www.egroups.com - Simplifying group communications

----- Original Message -----
From: <TomLeGeyt@...>
To: <shavian@...>
Sent: 17 December 1999 2:06 am
Subject: [shavian] Re: Homographs


> Lionel Ghoti submitted this puzzler:
>
> >The only problem I'm having with my Shavian transliterator Ghoti Filleter at
> the moment is with homographs.
>
> >To remedy this, I could either with (1); if not, then probably (2).
>
> >There can't be a huge amount, can there?
>
> I don't know. Who cares? Use solution #3.
>
> 3) Parse the words in pairs using surrounding words to determine the context,
> hence the pronounciation.
>
> examples:
>
> to live/ a live
> the wind/ to wind
> the present/ to present
> to lead/ of lead


But things aren't always as simple as that. Sometimes other parts of speech get in the way, and when they do, they're not always perfectly grammatical. E.g., "To really live", "A real live jackass". Instead of single words intervening, you could equally have whole phrases in their place, and eventually I think you'd end up having to parse the whole of the text, just to be sure.

On top of that, as I just said in a message to Scott Harrison, sometimes both pronunciations might have the same grammatical role, as in "lead" = dog leash vs. [pencil] "lead" = graphite: both are nouns.


> You won't need to:
> >1) add an extra field
>
> or:
> >2) build in a utility
>
> >Are there any homographs that have more than two possible pronunciations?
>
> If there are, this will solve them all. I would also suggest that writing the
> Shavian Transliterator in a language that will generate an executable file
> that will stand alone instead of a program that requires other software the
> end user may not have. You might even try an old version of BASIC. They are
> easy to aquire and inexpensive. (FREE)
>
> /tom leJet


------------------------------------------------------------------------
Don’t have the time or patience to search for unique
products & services? Then click to let us find them for you.
You Request, We Respond. http://clickhere.egroups.com/click/1957








-- Check out your group's private Chat room
-- http://www.egroups.com/ChatPage?listName=shavian&m=1

On 16 Dec 99, at 15:45, Scott Harrison wrote:

> In fact, one should be able to create HTML pages that are in UTF-8
> with mixed Shavian and any other alphabet on them. If so, we just
> need to determine how the browsers can be convinced to use specific
> Shavian fonts when encountering Shavian characters on a web page.
> Again, I am willing to help people be able to do this.

I think that what it comes down to is that we need fonts which have the
correct encoding -- i.e., which map the Unicode code point to the
correct glyph. At a guess, the current fonts (Lionspaw, Androcles etc.)
have the glyphs at code points corresponding to Latin letters. So we
would need someone to make a new font (or re-tool an old one) to
include both Latin letters and Shavian letters, with the correct
Unicode code point assignments. Then the browser might stand a fighting
chance of displaying the correct glyph.

Maybe someone could take a free Latin font and retrofit Shavian in the
ConScript Shavian range?

Cheers,
Philip
--
Philip Newton <Philip.Newton@...>

------------------------------------------------------------------------
Get 75% Off Our Best Selling Health and Wellness Books!
SelfCare.com has everything you need to take care of you and your family,
plus spend $40 or more and your shipping is free!
http://clickhere.egroups.com/click/1824



-- Easily schedule meetings and events using the group calendar!
-- http://www.egroups.com/cal?listname=shavian&m=1

Seems to me there is only one way to properly pronounce "really live"
and one way to pronounce "real live" no matter how grammar is involved.
Taken only two words at once, I don't think you'd end up having to parse
the whole of the text, just the two words.


>On top of that, as I just said in a message to Scott Harrison, sometimes
>both pronunciations might have the same grammatical role, as in
>"lead" = dog leash vs. [pencil] "lead" = graphite: both are nouns.

Unfortunately, I have to agree with you on this one.
"Sam bought his dog a new lead."
Can Sam's dog write or does he just need a new leash?

I'll keep thinking on this one.

/tom

------------------------------------------------------------------------
Get 75% Off Our Best Selling Health and Wellness Books!
SelfCare.com has everything you need to take care of you and your family,
plus spend $40 or more and your shipping is free!
http://clickhere.egroups.com/click/1824



-- 20 megs of disk space in your group's Document Vault
-- http://www.egroups.com/docvault/shavian/?m=1

The fonts on my web page have Shavian characters defined in the
Unicode space as well as the "normal" ASCII range. I believe the
Androcles and Ghoti fonts from Mr. DeMeyere also have the same
encoding, although I've never tried using the Unicode code points.

(My crude site is at www2.c4systm.com/~phild/ . I'm sorry, but the
Mac versions don't work.)

--Phillip Driscoll

-----Original Message-----
From: Philip Newton <Philip.Newton@...>
To: shavian@... <shavian@...>
Date: Friday, December 17, 1999 12:26 AM
Subject: [shavian] Re: Data formats for Shavian


>On 16 Dec 99, at 15:45, Scott Harrison wrote:
>
>> In fact, one should be able to create HTML pages that are in UTF-8
>> with mixed Shavian and any other alphabet on them. If so, we just
>> need to determine how the browsers can be convinced to use specific
>> Shavian fonts when encountering Shavian characters on a web page.
>> Again, I am willing to help people be able to do this.
>
>I think that what it comes down to is that we need fonts which have the
>correct encoding -- i.e., which map the Unicode code point to the
>correct glyph. At a guess, the current fonts (Lionspaw, Androcles etc.)
>have the glyphs at code points corresponding to Latin letters. So we
>would need someone to make a new font (or re-tool an old one) to
>include both Latin letters and Shavian letters, with the correct
>Unicode code point assignments. Then the browser might stand a fighting
>chance of displaying the correct glyph.
>
>Maybe someone could take a free Latin font and retrofit Shavian in the
>ConScript Shavian range?
>
>Cheers,
>Philip Newton <Philip.Newton@...>


------------------------------------------------------------------------
GRAB THE GATOR! FREE SOFTWARE DOES ALL THE TYPING FOR YOU!
Gator fills in forms and remembers passwords with NO TYPING at over
100,000 web sites! Get $100 in coupons for trying Gator!
http://clickhere.egroups.com/click/2092

-- Talk to your group with your own voice!
-- http://www.egroups.com/VoiceChatPage?listName=shavian&m=1

"Seems to me that George is a really live wire!"

"To really live one must learn something new each day."


-----Original Message-----
From: TomLeGeyt@... <TomLeGeyt@...>
To: shavian@... <shavian@...>
Date: Friday, December 17, 1999 12:43 AM
Subject: [shavian] Re: Homographs


>Seems to me there is only one way to properly pronounce "really live"
>and one way to pronounce "real live" no matter how grammar is involved.
>Taken only two words at once, I don't think you'd end up having to parse
>the whole of the text, just the two words.

>
>/tom


------------------------------------------------------------------------
FreeLotto.com is the only internet site to offer three free chances
a day to win $1,000,000.00. We've already had two $1,000,000.00
jackpot winners and thousands of other cash prizes. You could be
a $1,000,000.00 winner tonight! http://clickhere.egroups.com/click/2193

-- Create a poll/survey for your group!
-- http://www.egroups.com/vote?listname=shavian&m=1

TomLeGeyt@... wrote:
> Seems to me there is only one way to properly pronounce "really live"
> and one way to pronounce "real live" no matter how grammar is
> involved.
> Taken only two words at once, I don't think you'd end up
> having to parse the whole of the text, just the two words.

"That's an invalid remark, it must have come from an invalid."
(HAt's an *invAlid* ramRk, it must hAv kum from an *invalid*)

Cheers,
Philip

------------------------------------------------------------------------
GRAB THE GATOR! FREE SOFTWARE DOES ALL THE TYPING FOR YOU!
Gator fills in forms and remembers passwords with NO TYPING at over
100,000 web sites! Get $100 in coupons for trying Gator!
http://clickhere.egroups.com/click/2092

eGroups.com Home: http://www.egroups.com/group/shavian/
http://www.egroups.com - Simplifying group communications

Phillip Driscoll wrote:
> The fonts on my web page have Shavian characters defined in the
> Unicode space as well as the "normal" ASCII range.

Is that "Shaw Britannia", "Shaw Sans No. 1" etc.?

The following Shavian fonts on my NT machine show characters in the range
U+E700 - U+E72F (as per
http://www.indigo.ie/egt/standards/csur/shavian.html):

Shaw Britannia, Shaw Roman No. 1, Shaw Sans No. 1, Shaw Sans No. 2, Shaw
Sans No. 3, ShawCurly.

The following have no characters assigned in that range:

Androcles, Ghoti, Lionspaw, Shaw, ShawScript.

> I believe the
> Androcles and Ghoti fonts from Mr. DeMeyere also have the same
> encoding, although I've never tried using the Unicode code points.

The versions I have don't have anything at the "proper" Unicode code points,
but I may not have the newest version of the font.

> (My crude site is at www2.c4systm.com/~phild/ . I'm sorry, but the
> Mac versions don't work.)

Did you add this site to the links page at eGroups?

Cheers,
Philip

------------------------------------------------------------------------
Get 75% Off Our Best Selling Health and Wellness Books!
SelfCare.com has everything you need to take care of you and your family,
plus spend $40 or more and your shipping is free!
http://clickhere.egroups.com/click/1824



-- Check out your group's private Chat room
-- http://www.egroups.com/ChatPage?listName=shavian&m=1

Evenin' All

Here is my humble attempt at a transliteration.

Note this is not my accent. I cheated a bit by using TPT. (Well, I had to. I could hardly write a program and then not use it. That's a bit like going to a restaurant and at the end of the evening find that none of the staff eat there...) Anyway, I think the limitations of the CMU Dictionary are becoming quite clear to anyone using TPT. My argument is that I'm busy writing the new TPT for Windows so I have to cut corners where I can. :-)

So, without further ado, "The Homecoming":


"Hu hOmkumiN" (©) kopIrFt 1982, 1991 /A./m./kAluwE


Hu hOmkumiN

/hons sAt un Hu dOm rMm And lUkt Qt At Hu sIn. frum His vAntuJ pqnt hI kUd sI Hu Qtx limut uv Hu wxld: u flAt, blAk hxFzun wiH fYr kolumz, wun on Ic sFd.
tudE hI hAd u fIliN uv fUlfilmunt, uv destunI. Hu vFbrESunz hAd stortud, And Hu pqnts uv lFt ubuv Hu hxFzun hAd Siftud. sMn HE wUd xFv At Her nM wxld. hI wundxd weHx it woz kunektud wiH Hu tM uprOciN sfirz HAt hAd nQ kum intM vjM. hI'd trFd AskiN /kumpjMtx, but hI wozn't spIkiN tudE; uperuntlI sumwun hAd sed Hu rYN TiN tM him And hId gYn intM u sulk. jM hAd tM spIk tM /kumpjMtx un u verI orkEik stFl, Yr hI wUdn't Ansx, Yr giv jM funI Ansxz. hAns' frend /Jon sed pIpul spOk lFk him 500 jirz ugO.
/hons went bAk tM lUkiN Qt. Qt Her un Hu iNkI blAk wx TQzundz uv tFnI pqnts uv lFt, sImiNlI An infinutI uwE, solud And uncEnJiN. hI Yfun kEm up hir tM woc; Hu wxld Qt Her sImd sO vAst And emptI kumperd tM His krAmpt wYrun uv plAstik And metul; pIpul evrIwer And rubiS flOtiN ubQt. Qt Her woz dork And pIsful, hir woz brFtlI lit wiH flikxiN lFts, nqzI pIpul busuliN ubQt, /kumpjMtx bAbuliN uwE ubQt sumTiN. /hons felt Her destunI lE Qt Her, un HAt mistirIus wxld uv blAk.
hI'd YlwEz bin fAsunEtud bF it Az u cFld, And spent menI lYN Qxz un wun uv Hu dOmz, Just wociN. Aftx u fjM jirz hId bikum uwer HAt it woz slOlI cEnJiN: u stor HAt sAt Just ubuv Hu "hxFzun" mMvd klOsx And klOsx tM it until wun dE it tuct, Hen ivencuwulI it disupird YltugeHx. unuHx stor, dxektlI un frunt uv Hu dOm, hAd bin grOiN stedulI un brFtnus, sO muc sO HAt it nQ QtSOn evrITiN els, And bF its lFt hons kUd mEk Qt fIcxz on Hu QtsFd uv Hu wxld, And on Hu kolumz.
Hu vFbrESunz hAd stortud ubQt u wIk ugO, wFl evrIwun woz uslIp. TiNz HAt wx flOtiN ubQt lMs krASt intM wYlz, wYtx woz spild And uHx mFnx mishAps ukxd. wen pIpul uwOk Hu nekst dE Her wxld hAd sImd tM kum ulFv. Hu Encunt miSInxI woz fleksiN its musulz, redI fYr Hu grEt mOmunt HAt woz YlmOst upon Hu wxld. /kumpjMtx, Just bufYr hI went intM u sulk, tOld Hem HAt HE wx tM prIper Hemselvz fYr Hu /nM /wxld. hI menSund trIz, lEks, sunSFn And sumTiN kYld "grAvutI" wic jM kUdunt tuc Yr sI, but pUld evrITiN un wun dxekSun. nObodI kUd undxstAnd wut /kumpjMtx woz tYkiN ubQt, Ivun Hu eldxz. wen /hons went tM Hu dOm hI woz sxprFzd, Ivun Sokt, fYr Hu fxst tFm un hiz lFf Hu sIn QtsFd hAd cEnJd: Hu brFt stor woz nO lYNgx ubuv hiz hed but Ovx tM wun sFd, And Hu tM dimx obJekts hAd tEkun its plEs.

******************

/hons hEtud Hu "grAvutI" frum Hu mOmunt hI ikspirIunst it. wYtx nO lYNgx huN un sfirz wen jM pumpt it Qt uv its kuntEnx tM driNk it, but insted Sot uwE ukrYs Hu rMm And spred Ovx Hu wYl. pIpul kUd nO lYNgx mMv frIlI ubQt Hu wxld but kluN tM Hu wYlz. mMvmunt woz difukult; sum pIpul trFd slFdiN on Her bAks, uHxz on Her stumuks. /hons trFd Hu bAk meTud, but hI kUd not sI wer hI woz gOiN. sO wen hI xFvd At u hAcwE, hI did not sI it, lYst hiz bAluns And Sot strEt TrM it, hitiN Hu opuzut wYl rAHx vFuluntlI. /kumpjMtx tOld him Aftxwxdz HAt His SMtiN ubQt woz nOn Az "foliN" And woz tM bI uvqdud, but pIpul wx YlredI fFndiN His Qt. /kumpjMtx sed HE SUd pUS Her tYrsOz uwE frum Hu wYl, sO HE wx ruzistiN Hu fYrs wiH Her hAndz And nIz. His woz nOn Az "krYliN".
Hu fxst mYrniN wiH Hu grAvutI, wen Hu sFrun went Yf tM wEk evrIbodI, HE Yl Jumpt Qt uv bed un Hu nYrmul fASun, And ikspirIunst u sYrt uv "kumjMnutI fol". Hu nqz woz terubul. sum pIpul wx not un bed, but slept flOtiN un Hu midul uv Hu rMm. HIz pIpul Sot ukrYs Hu rMm wen Hu grAvutI stortud, OnlI tM hAv evrIwun els krASiN intM Hem wen Hu sFrun went Yf. nuTiN woz sEf. Yl jM hAd tM dM woz fxget wen jM let gO uv An obJekt, And uwE it went. uv kYrs jM hAd tM lUk Qt fYr uHx pIpulz flFiN obJekts.
/hons hAd mAnuJd tM strugul tM Hu dOm. Hu lAdxz HAt wx fitud un sum uv Hu pAsiJwEz wx nQ usenSul. bufYr Hu grAvutI HE wx simplI u hAndI TiN tM hOld wen munUvDiN ubQt. nQ HE wx AbsulMtlI usenSul fYr mMviN "up" And "dQn" Az /kumpjMtx pUt it. it woz YlsO kjUrIus, /hons nOtud, HAt Hu lAdxz igzistud OnlI un pAsuJuz wer HE wx nIdud. un sum pAsuJuz wun kUd not hAv jMzd lAdxz, And un suc plEsuz Her wx nun. HAt woz intrustiN. Hu lAdxz wx pUt Her wen Hu wxld woz bilt, hundrudz uv jirz ugO, And jet HE sImd speSulI plEst tM Ed Hu jMzx ugenst grAvutI, wic hAz OnlI Just stortud un Hu lAst fjM dEz. it woz YlmOst Az HO...
/hons YlwEz trFd tM mMv hed fxst, but wen trAvuliN "dQn" ulYN u lAdx, His wozunt suc u gUd FdIu Az hI jMZuwulI endud up lMziN hiz fUtiN And hitiN Hu "flYr" (His grAvutI woz OpuniN up u hOl nM vOkAbjulerI!). hQevx hI wYntud tM sI weHx Hu nM fYrs woz hAviN enI ufekt QtsFd, weHx Yl HOz obJekts wx YlsO "foliN dQn". /kumpjMtx tOld him HAt Hu grAvutI woz OnlI wuHin Hu wxld, And did not ufekt Hu QtsFd. hI YlsO sed HAt Hu grAvutI woz dM tM Hu "brEkiN prusIJx". HAt mEd sens, /hons TYt. hI'd nOtust u lot uv brEkiN wiH Hu grAvutI.
/hons rIct Hu dOm rMm And entxd it, Yr rAHx fel intM it, Az it woz nQ fEsiN dQn. un fAkt, it woz Hu mOst "dQn" port uv Hu wxld. Hu fAkt HAt it woz hAf fUl uv hot wYtx, un wic hons woz nQ sitiN, woz hordlI nOtusubul, kumperd tM Hu osum bjMtI QtsFd, wic hI woz nQ witnusiN. Qt uv Ic uv Hu fYr kolumz spxtud Jets uv blM-wFt lFt, briljuntlI lFtiN Hu dOm rMm And Qtx hul. TrMQt Hu sIn wx tFnI, wFt, flFiN portukulz, And Hu hOl sIn sImd tM Simx And wobul. And TrM Yl His, strEt dQn TrM Hu dOm, woz Hu obJekt, Hu bjMtuful, blM-grIn sfir HAt /hons nM must bI Her nM wxld. it huN Her, sFluntlI bekuniN And sumhQ, /hons TYt, rImOtlI fumiljx.

******************

it hAd bin u hord wIk, sins Hu grAvutI stortud. /kumpjMtx hAd bin tIciN Hem tM "wok", Hu impYrtuns uv up And dQn, And wF Hu "sIliN" woz nO lYNgx redulI Aksesubul tM Hem. hI'd YlsO tYt Hem ubQt freS er And rEn, hQ tM kIp wYrm And hQ tM plAnt TiNz. it Yl sImd verI kunfjMziN but, /kumpjMtx uSUrd Hem, it wUd Yl mEk sens un tFm.
/hons woz lFiN on hiz bed, redI tM tEk hiz slIpiN pil. HE Yl hAd tM strAp Hemselvz un, And tEk Hu pilz sO HAt HE slept TrM wut /kumpjMtx kYld Hu "rE-entrI pirIud". hI lE Her un AntisupESun, Hen Hu signul woz givun. hI swolOd hiz tAblut, And lE bAk. wen hI uwOk, hI TYt, hI wUd bI un Hu nM wxld, Hu wxld uv stAndiN And wYkiN And foliN dQn. Hu wxld uv freS er And rEn, uv sunSFn, trIz And plAntiN TiNz. hI driftud tM slIp.
hI uwOk sum Qxz lEtx, And Hu wxld woz sFlunt. Hu grAvutI woz stil Her, strYNx nQ, hI nOtust wiH unquns, but Hu vFbrESunz hAd stopt. hI struguld Qt uv bed, slOlI, tEkiN sum minuts, And slOlI mEd hiz wE dQn tM Hu usemblI erIu. u lorJ grMp hAd YlredI gAHxd Her, And wx sitiN Yr lFiN ubQt. Hu eldxz wx Her prIperiN tM Opun Hu hAc. wen mOst pIpul hAd xFvd, HE stortud txniN Hu big wIlz on Hu dYr; wIlz HAt hAd rumEnd unmMvd fYr probublI hAf u milenIum. Her woz u his uv er, And slOlI, ruluktuntlI At fxst, Hu big dYr Opund.
Hu inhAbutunts uv Hu Old wxld stept Qt intM Hu xlI mYrniN uv Hu nM wxld. it woz stil YlmOst dork, but uwE un wun dxekSun Her woz u blM-grE glO. HE kUd Just mEk Qt dutElz xQnd Hem frum Hu Tin but grOiN lFt, wic sImd tM mimik Hu pEl lFt wic slOlI streNTund And ilMmunEtud /hAnz' dOm rMm sIn. evrIbodI woz lUkiN wiH umEzmunt tuwYrdz Hu uprOciN dYn. it sImd sO for, for uwE. sMn uHx fIcxz stortud tM utrAkt Her utenSun. Her wxld hAd lAndud smAk un Hu midul uv sum sYrt uv plAtfYrm, YlmOst Az HO dulibxutlI, And Yl xQnd HE kUd disxn SEps. sumwer un Hu dorknus sumTiN woz mEkiN u sYrt uv stutxd wisliN nqz, And /hons wundxd wic wx Hu trIz And TiNz HAt nIdud tM bI plAntud. sunSFn woz lFt, sO HAts wut Hu glO must bI, but hI kUdn't wxk Qt wer Hu rEn woz.
wFl His woz gOiN on sum uHx SEps hAd bin mMviN slOlI tuwYrdz Hem, And Hu lFt woz grOiN brFt unuf sO HE kUd sI HAt Hu SEps wx pIpul. His wxld hAd pIpul on it! TrI uv Hem nQ mMvd fYrwxd wiH pxpus, until Ic grMp wx Ebul tM sI Hu uHxz' fEsuz. HIz nM wxld inhAbutunts wYr kYrs gormunts uv leHx And ruf wOvun mutirIul, And sum wYr u litul ormx. HE wx SYrt, stokI And hevI, probublI frum u lFftFm uv ruzistiN grAvutI. HE ubzxvd Hu strEnJ vizutxz un Hu pEl lFt frum wuHin Hu hAcwE, And frum Hu grOiN dYn. pjMnI, skinI And pEl, hM wx HIz strEnJ pIpul hM kept foliN dQn, skAntulI klAd un strEnJ gormunts? Hu nEtivz hAd sIn His TiN kum TundxiN dQn, And ruSt tM sI. hir it woz, Just Az Hu leJundz sed it wUd bI, And Hen HIz funI pIpul kEm Qt.
HE stUd Her fYr u litul wFl, rugordiN wun unuHx un Hu mYrniN lFt, Hen wun eldx stept fYrwxd, mEntEniN hiz bAluns, hiz orm Qtstrect Az /kumpjMtx hAd uperuntlI tOld him, And wun uv Hu TrI nEtivz tUk hiz hAnd wiH u grip uv Fxn. Hen Hu uHx tM mMvd TrM Hu TrYN, SEkiN hAndz And smFliN, And Hu rest uv Hu nEtiv grMp mMvd fYrwxd And did lFkwFz. Hu trAvulxz nM HAt HE wx welkum hir. And Az Hu sun HAt /hons hAd woct fYr sO lYN rOz Ovx Hu mQntunz And SOn dQn upon Hem, HE nM HAt HE wx hOm. HE hAd rutxnd tM /plAnut /xT.

I hope you enjoyed it. I will post it to the Vault presently.


- .+'^'+. A.M.Callaway ----------------- acal@...
- A N D Y Melbourne, Australia --- a.callaway@...
- `+.,.+' www.ozemail.com.au/~acal <http://www.ozemail.com.au/~acal> -------------------------
________________________________

Get cash for you...and you...and you... <http://clickhere.egroups.com/click/2238>
Get $20 for signing up, then get $10 when each of your friends do.

<http://clickhere.egroups.com/img/2238/xlogo.gif>
eGroups.com Home: http://www.egroups.com/group/shavian
www.egroups.com <http://www.egroups.com> - Simplifying group communications