[E-trademarks] Unicode Growing Pains - Recent Registration Certificate Examples Are Disappointing

Wed Apr 16 21:32:00 UTC 2025

Some additional observations.

Search the term Unicode on the TMEP and what do you get?  Nothing. Same for searching unicode on the USPTO website with pages/documents restricted to Trademarks guidance.

I somehow got the DM - Decimal Mark search field on TESS.  That DM field allowed for searching any particular character in wordmark entries by the decimal value of the character.  Trademark Search does NOT have a similar search field, or am I mistaken?   Finding a particular character in wordmark entries relies on the search engine, and the search engine has some querks.  For example, the search WD:*ß* AND LD:true to find all occurrences of the small sharp s, sz ligature character returns 150,963 trademarks, but the hits include marks like BAD ASS ACADEMY, PAPERLESS SHOP, and TOSS ~ N ~ TOWEL GAME TOWEL, basically including all wordmarks with two consecutive occurrences of the character S, so finding actual occurrences of ß (which likely do occur) is challenging.  More to the point, searching for Unicode characters with decimal values above 382 in wordmark entries (or any other search field) is challenging (assuming you're not interested searching the thousands of Unicode characters one at a time).

The search WD:*µ* AND LD:true AND MD:standard, where µ is the micro sign (decimal 181) straight from the Standard Character Set, retrieves 51 trademarks, namely (decreasing SN order):

#
SN
FD
Status
Wordmark
1
98278476
11/20/23
Pend
µICROTHORACOTOMY
2
98102625
07/26/23
Pend
AµTOPULSE
3
97914820
05/01/23
Pend
µDOSE
4
97762585
01/20/23
Pend
µCAB
5
97559973
08/23/22
Reg
STRATµM
6
97524133
07/28/22
Pend
µ REPAIR
7
97457377
06/14/22
Pend
AµSTREAMER
8
97280279
02/23/22
Reg
µASIC
9
97168849
12/13/21
Pend
µLIF
10
90866614
08/05/21
Reg
µ-PIXELED
11
90481928
01/22/21
Reg
µ-VISOR
12
90010307
06/19/20
Reg
µHARVEST
13
90002090
06/15/20
Reg
µMASK
14
88778209
01/29/20
Reg
WAVEGUIDE FORMµLA
15
88584251
08/19/19
Reg
µPULSE
16
88438880
05/20/19
Reg
µFINEX
17
88438874
05/20/19
Reg
µFINEX
18
88438863
05/20/19
Reg
µFINEX
19
88266018
01/17/19
Reg
µFILMTEC
20
88177708
11/01/18
Reg
PICO TOµCH
21
87885838
04/20/18
Reg
VµLD
22
87885796
04/20/18
Reg
FµLD
23
87829914
03/12/18
Reg
µDOO
24
87820929
03/05/18
Reg
µHASH CONTRACT
25
87703251
11/30/17
Reg
µP
26
87569030
08/15/17
Reg
PICO PµLSE
27
87475127
06/05/17
Reg
BµG
28
87292043
01/06/17
Reg
µ-SPOT
29
87212858
10/24/16
Reg
OMNIA µMPX
30
87030714
05/10/16
Reg
LE-µ'S
31
86920852
02/26/16
Reg
µHX
32
86844024
12/09/15
Reg
µELUTION
33
86780956
10/07/15
Reg
A-ZOOMµ
34
86495252
01/05/15
Reg
HµREL
35
86383746
09/03/14
Reg
OPTIMµM
36
86170676
01/21/14
Reg
DµRT HOG
37
86116640
11/12/13
Reg
µNID
38
85859222
02/25/13
Reg
EPPENDORF µCUVETTE
39
85677815
07/16/12
Reg
µMODULE
40
85085675
07/15/10
Reg
µPOD
41
79326919
08/12/21
Reg
µGLED
42
79303058
12/07/20
Reg
µSPEAKER
43
79215859
06/12/17
Reg
MµFOCUS
44
79084190
05/14/10
Reg
µSPRINT
45
79066115
02/13/09
Reg
µ TASWAKO
46
79037473
01/19/07
Reg
µWAVE WIZARD
47
78600165
04/01/05
Reg
µPA
48
78543309
01/06/05
Reg
µARRAY
49
77634350
12/16/08
Reg
µEZ
50
77567600
09/11/08
Reg
µQC
51
77245444
08/02/07
Reg
µBRICKS

Conversely, the search WD:*μ* AND LD:true AND MD:standard retrieves 9 pending trademarks, namely (decreasing SN order):

#
SN
FD
Status
Wordmark
1
99109969
03/28/25
Pend
IMΜNOGEL
2
98859737
11/18/24
Pend
ΜPNT
3
98761636
09/20/24
Pend
Μ - MATERIAL MACHINE
4
98671363
07/29/24
Pend
ΜFIDELITY
5
98671353
07/29/24
Pend
ΜVALVE
6
98671333
07/29/24
Pend
ΜCOOLING
7
98278543
11/20/23
Pend
ΜTHORACOTOMY
8
98168957
09/07/23
Pend
ΜMIST
9
97836045
03/13/23
Pend
ΜDRIVE NEXUS

The last 3 are highlighted as they have published for opposition date entries.

The difference is that the second μ search uses the Unicode character with decimal value 956, plus USPTO's wordmark processing converted that lower case μ to the upper case Μ (decimal 924) that visually matches the ordinary upper case M. I'm guessing every character in the USPTO standard character set has an equivalent Unicode character with a decimal value over 382, the highest decimal value in the standard character set. (I've noticed several others, but no since one else is sharing, …)

Back in the early 2000's, probably around the time the standard character set was defined (before the US joined the Madrid Protocol), I met with the OCIO to discuss/define the character equivalents for X-Search/TESS.  Most of the equivalents were obvious (e.g., lower case a = upper case A = A with any diacritical mark), but there were a few exceptions.  I think we equated characters like Æ to the first letter of the merged pair.  As I recall, I excluded ƒ (small italic f, function of, - decimal 131) from being equated to lower case f = upper case F.  I wanted the few occurrences of ƒ to be uniquely found by searching ƒ. Same for ß (small sharp s, sz ligature - decimal 223).  I rejected equating ß to S, and TESS didn't allow equating ß to the SS letter pair, so ß was not equated to any other character on TESS.  Well, that's not the case on Trademark Search, as ß has been equated to the SS letter pair, making identifying actual occurrences of ß on Trademark Search challenging among the SS records returned for the ß search. Oh … I already mentioned that at the start of this email. So it goes.  (All this was done long before Trademarks management authorized my request to add the DM - Decimal Mark index.  Then again, I'm guessing that I was among a small list of TESS users who regularly used the DM search field - that if lots of searchers were using the DM search field, the USPTO would have been reluctant to exclude that search field from Trademark Search, right?  So it goes.)

Happy Unicoding!¿¡?!
Ken Boone

________________________________
From: Carl Oppedahl <carl at oppedahl.com>
Sent: Wednesday, April 16, 2025 4:00 AM
To: For trademark practitioners. This is not for laypersons to seek legal advice. <e-trademarks at oppedahl-lists.com>
Subject: Re: [E-trademarks] Unicode Growing Pains - Recent Registration Certificate Examples Are Disappointing

Thank you Ken for posting.

I have several reactions to this.

First, Ken's postings in the past few months about character coding have prompted me personally to try to learn about Unicode.  It is a fascinating world, this Unicode.  As time goes on, I must imagine that trademark offices around the world will eventually gain familiarity with Unicode.  The result, eventually, will be better ways of searching, and better ways of storing mark information for searching.  And better ways of receiving trademark applications in the first place, with applicants providing Unicode representations of marks rather than mere images of marks.

Second, Ken's postings have laid bare the many ways in which Trademark Center (and USPTO's other related internal systems for trademark application workflow) have failed to keep pace with Unicode.  Yes it is one thing if, within recent days, the USPTO coders belatedly started to check to see if the "mark" field in an application is or is not composed solely of "standard characters".  But it is clear the USPTO coders have not been checking to see if other fields (such as the "translation" field and mailing address) contain non-ASCII Unicode characters.

WIPO, as the administrator of the PCT, Madrid, and Hague systems, has historically served the IP community by nudging the world's intellectual property offices along towards current developments.  There are many examples of this.  See for example the ST.26 standard for submission of computer-readable genetic sequence listings.  I have to imagine that our friends at WIPO are trying, as best they can, to think about Unicode.  One of the challenges, of course, is that because of the way that Madrid Protocol is structured, nobody can file a Madrid application directly at the International Bureau.  (Direct filing at the IB is possible for PCT and Hague applications, but not for Madrid applications.)  Instead, the only entry path for a Madrid application is a filing in one or another of the Offices of Origin.  OoOs surely differ greatly from one to the next as to the richness or paucity of the various data fields.  For all I know there may be some OoOs for which the filing path even now in 2025 is in the nature of stone tablets with chiseled writing.

Some trademark offices are in places where non-Latin (non-ASCII) characters are very important.  Here you can see (WIPO web site<https://www.wipo.int/en/ipfactsandfigures/trademarks>) a ranking of the ten biggest users of the Madrid system:

  1.  US
  2.  Germany
  3.  China
  4.  France
  5.  UK
  6.  Switzerland
  7.  Japan
  8.  Italy
  9.  Korea
  10. Australia

China, Japan, and Korea show up in the top ten, and in each of those places, non-Latin (non-ASCII) characters are very important.  My sense is that Unicode by now supports most languages including Chinese (simplified and traditional), Japanese (katakana, hirigana, and kanji), and Korean (hangul).  Hopefully eventually the highest-volume trademark offices will get together to try to work out ways to make use of Unicode for filing of trademark applications, for searching, for publication, and for other workflow purposes. Hopefully eventually it would reach the point where a Madrid filing could contain a Unicode mark, and no matter which Office is designated, the IB could transmit the designation to the designated Office and that Office could actually know what to do with the Unicode characters.

On 4/15/2025 2:40 PM, Ken Boone via E-trademarks wrote:
Following are 4 recent registrations with Unicode characters in the translation/transliteration fields (as evident on the TSDR summary tab).  In each case, the Registration Certificate simply dropped the Unicode characters.

Drawing
SN
RD
Comment
 [previously viewed                          Image for 98496184, select for more details]

98496184
01/21/25
 Translation:  The wording Benbo奔博 has no meaning in a foreign language.

Reg. Cert.:    The wording Benbo has no meaning in a foreign language.

 [previously viewed                          Image for 98469783, select for more details]

98469783
04/08/25
 Translation: The English translation of 雪冰 in the mark is SNOW ICE.

Reg. Cert.:   The English translation of  in the mark is SNOW ICE.

 [previously viewed                          Image for 98384496, select for more details]

98384496
02/11/25
 Translation:  The English translation of 金满庭 in the mark is Gold Filled Palace.

Reg. Cert.:  The English translation of  in the mark is Gold Filled Palace.

 [previously viewed                          Image for 98018070, select for more details]

98018070
02/04/25
 Translation:  The English translation of 瑞安房地產 in the mark is auspicious; peaceful; real estate.

Reg. Cert.:  The English translation of  in the mark is auspicious; peaceful; real estate.

Happy Unicoding!?!?!
Ken Boone

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oppedahl-lists.com/pipermail/e-trademarks_oppedahl-lists.com/attachments/20250416/c8994116/attachment.html>