[E-trademarks] TEAS Voluntary Amendment Doesn't Like Unicode Characters - See 99414215 - ?OCCETTA ????? RE INVENTED

[Ken Boone]

Appending to Carl's comments about Unicode characters in trademarks:

By my checks, the term Unicode does NOT appear in the TMEP. A similar search of the USPTO website, accepting Web pages and/or Documents for Trademarks guidance fails to retrieve any results.  More bluntly, the USPTO currently lacks specific guidance on Unicode characters.  (As I recall, the online help for edit tool TradeUps at the USPTO provided detailed guidance for wordmark entries. Alas, I do not have a copy of that guidance.  Then again, that guidance was written before the USPTO entered the Madrid Protocol and was often ignored.)

I have witnessed multiple examples of Pre-Exam removing Unicode characters from wordmark entries, typically for Arabic characters or Asian characters.  Some examples include the following.
SN
Current Wordmark
Original Wordmark
Drawing
99276699

ピット
 [Image for 99276699]
99197989
ABC YOUTH EQUIPPING EDUCATING AND EMPOWERING YOUNG MINDS
' عد EQUIPPING EDUCATING AND EMPOWERING YOUNG MINDS YOUTH
 [Image for 99197989]
99162243
OMAKASE BY OSEN AUTHENTIC JAPANESE RESTAURANT
おせん (JAPANESE CHARACTERS, MEANING OSEN IN ENGLISH )OMAKASE BY OSEN AUTHENTIC JAPANESE RESTAURANT
 [Image for 99162243]
99160404

ARABIC CHARACTERS دار الورد
 [Image for 99160404]
99147291
KOCHI
كز KOCHI
 [Image for 99147291]
 I witnessed similar behavior by Pre-Exam regarding standard character marks, namely that Pre-Exam would change the mark drawing code from standard characters to text & design to add the design code for < or >, as the trademark editor would not allow design codes to be entered for standard character marks.  Some likely examples:  97647751 FLAW<LESS;  97750213 we<others; 97607290 E>x<ecute.

Technically, for non-Latin characters, design codes are appropriate, except at least one problem is that Pre-Exam may not be aware that non-Latin character occur in wordmark entries.  Here's a brief list of some relevant design codes.


28.01.01 Inscriptions in Arabic characters

28.01.02 Southeast Asian characters, e.g. Thai, Khmer, Indian languages (e.g. Hindi, Punjabi, Bengali, Gujarat)

28.01.03 Japanese and Chinese characters Formerly Inscriptions in Chinese, Japanese, Korean, Vietnamese or other Asian characters

28.01.04 Korean Hangul

28.01.05 Inscriptions in Greek and Cyrillic characters Formerly Inscriptions in Greek characters

Pi symbols which are not accompanied by other Greek characters are in 24.17.11 exclusively. Pi symbols with other Greek letters are in 28.01.05 .

28.01.07 Inscriptions in Hebrew characters, including the following: ת ש ר ק צ ץ פ ף ע ס נ ן מ ם ל כ ך י ט ח ז ו ה ד ג ב א

28.01.25 Inscriptions in other non-Latin characters, including hieroglyphic characters Formerly Inscriptions in other non-Latin characters, including Cyrillic (Russian) or hieroglyphic characters

If the USPTO has a tool for identifying Unicode characters that are not what they seem (like В - the CYRILLIC CAPITAL LETTER VE that strongly resembles the standard character B), that would be news to me.

Given that characters outside the standard character set are often difficult to recognize visually, I recommend that some applications (say TSDR and Trademark Search) be updated to identify any characters in wordmark entries that are outside the standard character set regardless of the mark drawing code for the trademark.  Following are some sample standard character marks where I highlighted the challenging characters (except for CAFÉ  NAIROBI, the challenging character is not printable, so this highlighting feature was not successful --- that some dump utility like Carl's favorite character dumper (https://www.babelstone.co.uk/Unicode/whatisit.html) would be a more appropriate tool, assuming the USPTO folks could duplicate the results in a more compact format).  I leave the task of using Carl's favorite character dumper for these trademarks to the user.

99370606    ΜTUMA
99341607    ALLUBI اللوّبي
99292236    AI ≠ I AM
99283545    ​RAPIDREST
99282586    HOʻI
99265212    CAFÉ  NAIROBI
99135001    Z‧PILOT
98492294    KῙNᾹ'OLE
98111449    KYŪ​​​​​​​R
79425207    FIX² TURBO GT

For your convenience, the search SN:( 99370606 99341607 99292236 99283545 99282586 99265212 99135001 98492294 98111449 79425207 ) retrieves these 10 trademarks.

BTW, I use the unicode(), unichar(), code() and char() functions in Excel to analyze characters.  I still miss the DM - Decimal Mark search field of TESS.  While I was at the USPTO, I could access TESS in Native Mode and view/export the contents of the DM - Decimal Mark index for any trademark on TESS.  In Trademark Search syntax, the search LD:true AND DM:/<256-9999999>/ likely would retrieve all live marks with any Unicode character with decimal value greater than 255 in the wordmark.

Happy Unicode Searching,
Ken Boone

________________________________
From: Carl Oppedahl <carl at oppedahl.com>
Sent: Monday, October 6, 2025 1:03 PM
To: For trademark practitioners. This is not for laypersons to seek legal advice. <e-trademarks at oppedahl-lists.com>
Subject: Re: [E-trademarks] TEAS Voluntary Amendment Doesn't Like Unicode Characters - See 99414215 - ?OCCETTA ????? RE INVENTED


Thank you Ken.

This is, of course, a pro se case.   And not that it's relevant to the present topic of USPTO failing to handle Unicode correctly, I have to imagine the Examiner is going to refuse to enter the amendment on the grounds that adding color is a material alteration.

Returning to the Unicode topic.

There have been several of these USPTO-failing-to-handle-Unicode-correctly cases where the seeming common theme is that the filer used one or more characters that are unnecessarily exotic.  Here, for example, the application as filed used the "U+0412 : CYRILLIC CAPITAL LETTER VE" which in terms of ink on the page looks almost exactly like the Latin or Roman (or ASCII) letter "B".

Sometimes when somebody uses (for example) a Cyrillic VE ("В") instead of a Latin B, it is because the person is trying to be sneaky and get away with something.  For example the sneaky person registered the Internet domain name "CITIВANK.COM" (using the Cyrillic VE) and is planning to try to use it to steal money from customers of the real Citibank by setting up a fake duplicate of the Citibank web site.

Here, I don't immediately see anything in the TSDR record that suggests the trademark applicant is trying to get away with something evil.

Well, I guess I could think of one thing an applicant might be up to here.

Let's suppose our applicant in this case had done a search of Office records and saw that somebody already has a strong trademark registration for a standard-character word mark that looks just like this pending mark (in terms of ink on the page).  And suppose the applicant is worried that that Examiner might impose a 2d refusal over that standard-character registration.  So then maybe the applicant hopes the Examiner will copy and paste the applied-for characters (including the Cyrillic VE) into the Examiner's search system and will  get no hits.  And thus no 2d refusal.

I don't mean to say that I have any particular reason to think the applicant is really trying to do this.  I am just trying to guess about things.

But then it seems to me that we would be reduced to trying to guess what innocent sequence of events would lead to this case having (for example) a Cyrillic VE ("В") instead of a Latin B.  Was this the result of a super-buggy OCR engine within TC?



On 10/6/2025 11:27 AM, Ken Boone via E-trademarks wrote:
Trademark Experts,

Caution: It appears that the TEAS Voluntary Amendment form does not understand Unicode characters outside the ASCII character set, substituting question marks (?) for such Unicode characters.

Consider 99414215, filed on 9/26 with В CCETTA ВОССЕ RE INVENTED appearing as the wordmark.

[Image for 99414215]

Today, ?OCCETTA ????? RE INVENTED is presented as the wordmark.  Why?  The Prosecution History shows that on Oct. 05, 2025, a TEAS VOLUNTARY AMENDMENT was received, amending the drawing (the original drawing did not include colors).  The new literal element ? CCETTA ????? RE INVENTED was provided. (I do not see any evidence that Pre-Exam has performed any edits of this new application yet {i.e, the prosecution history does not include the NEW APPLICATION OFFICE SUPPLIED DATA ENTERED entry}.)

Using Carl's favorite character dumper (https://www.babelstone.co.uk/Unicode/whatisit.html), the original wordmark В CCETTA ВОССЕ RE INVENTED has multiple Unicode characters.

U+0412 : CYRILLIC CAPITAL LETTER VE
U+0020 : SPACE [SP]
U+0043 : LATIN CAPITAL LETTER C
U+0043 : LATIN CAPITAL LETTER C
U+0045 : LATIN CAPITAL LETTER E
U+0054 : LATIN CAPITAL LETTER T
U+0054 : LATIN CAPITAL LETTER T
U+0041 : LATIN CAPITAL LETTER A
U+0020 : SPACE [SP]
U+0412 : CYRILLIC CAPITAL LETTER VE
U+041E : CYRILLIC CAPITAL LETTER O
U+0421 : CYRILLIC CAPITAL LETTER ES
U+0421 : CYRILLIC CAPITAL LETTER ES
U+0415 : CYRILLIC CAPITAL LETTER IE
U+0020 : SPACE [SP]
U+0052 : LATIN CAPITAL LETTER R
U+0045 : LATIN CAPITAL LETTER E
U+0020 : SPACE [SP]
U+0049 : LATIN CAPITAL LETTER I
U+004E : LATIN CAPITAL LETTER N
U+0056 : LATIN CAPITAL LETTER V
U+0045 : LATIN CAPITAL LETTER E
U+004E : LATIN CAPITAL LETTER N
U+0054 : LATIN CAPITAL LETTER T
U+0045 : LATIN CAPITAL LETTER E
U+0044 : LATIN CAPITAL LETTER D

I'm concluding that the newly provided literal element ? CCETTA ????? RE INVENTED closely matched the original wordmark ?OCCETTA ????? RE INVENTED, but that the TEAS VOLUNTARY AMENDMENT did not recognize the Unicode characters outside the ASCII character set.

BTW, everyone has noticed that the Beta viewer for Trademark Search still provides incomplete information, right?  The Announcements page still shows the notice Update: On September 24, we temporarily disabled some of these new beta features to test them further. We'll remove this message once they are restored.

Happy Trademark Amending,
Ken Boone

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oppedahl-lists.com/pipermail/e-trademarks_oppedahl-lists.com/attachments/20251006/b190eb5e/attachment.html>