it is just your correspondents actually.

The emails are most likely incorrectly encoded. I had discussion on this forum in the past few day with a chap who was wedded to Times New Roman font, so he encoded his emails in an old character set so he could do it his way. Unfortunately all the mail he sent will demonstrate the problem you have. He, I can assure you, does not care what his recipients see, only that he sees his beloved Times New Roman.

So to a workaround. View menu (alt+V) select text encoding and manually select Unicode for that email

I'm sorry to disagree with you Matt but it is not the fonts, I have excluded them by experimenting. The same font does not always give the same faults, most times they are fine. Neither is it always the same correspondents, sometimes they are fine from the same correspondents. The fault lies either in Thunderbird or in some other area of my computer and I don't know how to isolate it. Thanks for your help anyway

Did you try changing the character encoding, which is what is wrong with the emails?

You spent some time telling me I am wrong, but you never addressed the thing that will actually fix the issue.

I have dealt with hundred of people with the same issue. I tried to give you some background, but doing so apparently moved your eye from the ball. The issue is not Thunderbird. You will need to prove to me that it is anyway as this is a common issue with a common resolution.

I know that one can "fix" the e-mail by Alt V but that does not solve the problem as to why they are displayed the way they are. One shouldn't have to do that if Thunderbird was performing properly as the fonts are quite normal fonts and as I said it doesn't always happen with the same font or the same sender.

Thunderbird displays the email in the character set defined by the sender. So if the sender specifies something other than unicode then your results will be sub optimal. I still see lots of mail using old windows code pages. Most probably from people still using old windows PCs

To take the example of the email I received saying you had replied.

MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit

So Thunderbird displayed the message as unicode.

Now, as you know, fonts are not all created equal. what characters map to what locations above the basic keyboard characters is really not defined and in the case of the old code pages limits to 255 characters. Unicode has some 12,000 characters.

The common reasons for issues are the "smart" quotations and commas I will not try and explain that issue. Wikipedia does a pretty good job of it. https://en.wikipedia.org/wiki/Quotation_mark#Curved_quotes_and_Unicode

From my perspective it appear that those that care most about the presentation of their mail are those most likely to try and use typographical notations and therefore those most likely to send messed up messages.

Stall said

One shouldn't have to do that if Thunderbird was performing properly as the fonts are quite normal fonts and as I said it doesn't always happen with the same font or the same sender.

How it the email program supposed to guess that this mail was incorrectly formatted by it's sender and apply the appropriate change to force the email to display correctly.

Hi Matt, Thanks for your explanation. The two most frequent correspondents I have use Arial and Comic Sans SM which I think you will agree are basic and not old. One of those correspondents uses Thunderbird and though I do get mail from others they are the two I deal with most. Neither of those two have old PC's and both use windows 10 as I do. I do not have any problems with their mail on my Thunderbird on my laptop only on the desktop so there must be a problem with the desktop computer or the download of Thunderbird on that computer. I just do not know how to isolate the problem.

check if there is a language setting. configure to english.

No language setting in Thunderbird.

Check the character encoding in the emails source..... That has to be the first part so we do know if we are dealing with.

I did not mention this because it sounds like Microsoft bashing and is only another cause. But the truth of the matter is, many people compose their emails in Word and copy and paste to Thunderbird. Word forces a Windows code page by including it in the copy and paste.

But regardless, check the mails on your computer that does not do it as well. Just to be sure the code pages are the same. Also look in Options > Display > Advanced as to the default encoding. THat might shed some light.

Hi Matt. How do I check the character encoding in the e-mail source? What am I looking for? None of my correspondents compose their e-mails in word, they do them normally. I changed the default coding to Unicode some time ago and thought this had cured the problem but it came back after a while.

Select the message, press and hold ctrl, tap u. You'll also find it in the menu under View.

Thanks Zenos but I still don't know what I am looking for?

Something like Matt showed you earlier:

MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit

The "charset" is of particular interest. ctrl+f will bring up a search tool.

Hi Zenos, Here is a copy. It seems to me that the charset is 8pt unicade .X-Junkmail-Premium-Raw: score=7/50,refid=2.7.2:2016.11.13.141817:17:7.944,ip=212.159.14.19,rules=__CSHC_RCVD,

__TO_MALFORMED_2, __TO_NAME, __TO_NAME_DIFF_FROM_ACC, __HAS_FROM,
__FROM_HAS_AT, __SUBJ_ALPHA_END, __HAS_MSGID, __SANE_MSGID, __USER_AGENT,
__MOZILLA_USER_AGENT, __MIME_VERSION, __CT, __CTYPE_MULTIPART_ALT,
__CTYPE_HAS_BOUNDARY, __CTYPE_MULTIPART, __BAT_BOUNDARY, __MIME_TEXT_P2,
__MIME_TEXT_H2, __HIGHBITS, __C230066_P5, __HAS_HTML, HTML_NO_HTTP,
BODYTEXTP_SIZE_3000_LESS, BODY_SIZE_5000_5999, BODYTEXTH_SIZE_10000_LESS,
__MIME_TEXT_H1, __MIME_TEXT_P1, __MIME_HTML, __TAG_EXISTS_HTML, NO_URI_FOUND,
NO_CTA_URI_FOUND, __MIME_TEXT_H, __MIME_TEXT_P, BODY_SIZE_7000_LESS,
NO_URI_HTTPS, __TO_REAL_NAMES

Received: from avasout04.plus.net (212.159.14.19) by rgin08.bt.int.cpcloud.co.uk (9.0.019.12-1)

       id 5824DB3103E7E361 for [email protected]; Sun, 13 Nov 2016 15:11:37 +0000

Received: from [192.168.1.2] ([46.208.125.47]) by avasout04 with smtp id 7TBa1u00411UmE601TBcSq; Sun, 13 Nov 2016 15:11:36 +0000 X-CM-Score: 0.00 X-CNFS-Analysis: v=2.2 cv=T4LOdLCQ c=1 sm=1 tr=0

a=r5lHtdHmuq+hXG3cBJwfzg==:117 a=r5lHtdHmuq+hXG3cBJwfzg==:17
a=MKtGQD3n3ToA:10 a=dik7HMQjudEA:10 a=ZZnuYtJkoWoA:10
a=r77TgQKjGQsHNAKrUKIA:9 a=dD_2W9rf9TNBLG181KgA:9 a=QEXdDO2ut3YA:10
a=fWyW7b-8QGHVZLRO_3IA:9 a=jwrZzYDhKwogdGK6:21 a=_W_S_7VecoQA:10

X-AUTH: hporte@:2520 To: From: Subject: Memory checks Message-ID: <[email protected]> Date: Sun, 13 Nov 2016 15:11:34 +0000 User-Agent: Mozilla/5.0 (X11; Linux i686; rv:45.0) Gecko/20100101

Thunderbird/45.4.0

MIME-Version: 1.0 Content-Type: multipart/alternative;

boundary="------------818C28EADF8149FA3B712313"

This is a multi-part message in MIME format.

Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit

Hi Terry,

Please keep an open mind on Windows hardware diagnostics. You will, no doubt have had the experience '- as I have on several occasions - of Windows saying "This device is working properly" when your printer is refusing to work and is obviously faulty.

The average PC has about 4 Gigabytes of memory. ( 4000000000 bytes = 32000000000 bits). Addressing, writing to and reading back from each of these is a long process. At least in older computers, shortcuts tended to be taken, by testing large blocks at once. They did this by 'Parity checking' . This simply determined whether the contents of the block added up to an even number of 'bits' in the set or unset state. Often, just the end bit of a row was used, as you can see is possible from the example below.

• Dec Hex Binary Parity*

  15               7            0

Space 32 20 000000000000000000010100 even

     Ä          142         8E 000000000000000010001110            even

Note: Comparing the binary code for a space and Ä, bits seven, four, three and one are reversed. Bit 0 unaffected.

A memory chip will have millions of bits in locations arranged in a grid. A small faulty patch in that grid corresponding to the digits 7 down to 1 above could cause a space to be mistaken for Ä in the output text.

Note that the end bit (Bit 0) is not affected, so a memory check based on parity would pass that patch as normal. I know that Windows memory checks used to be simple parity checks, but do not know whether that is still so. You should be able to find out by Googling ‘Parity memory checks and Windows’. As memory gets ever bigger, the incentive to take shortcuts must be even greater.

So please keep an open mind.

Hubert.

Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 8bit

I'm not sure the above went?

Hi Matt and Zeros,

I just thought I would let you know that my problem now seems to be solved thanks. Eventually I had to completely uninstall Thunderbird and reinstall it from scratch as there must have been a problem with the original download which meant that the text code would not remain on Unicode and kept reverting to Western. It now stays on Unicode and all seems well.

Thank you for your help and support.

Stall said

Hi Matt and Zeros, I just thought I would let you know that my problem now seems to be solved thanks. Eventually I had to completely uninstall Thunderbird and reinstall it from scratch as there must have been a problem with the original download which meant that the text code would not remain on Unicode and kept reverting to Western. It now stays on Unicode and all seems well.Thank you for your help and support.