Series codes are very important. Anybody who regularly practices before the USPTO has more or less memorized the most commonly encountered series codes in his or her own daily work. If you see series code 62, you know it is a provisional application. If you see series code 29, you know it is a design application. Series code 91 means oppositions before the TTAB. Series code 92 means cancellation proceedings before the TTAB. Every now and then, a series code gets “used up” meaning that serial number “999,999” is reached in the series code, and then things roll over into a new and different series code. Provisionals, for example, used to be in series code 61 and before that they were in series code 60.
For over fifty years, for as long as there have been series codes at all, the trademark folks and the patent folks have carefully avoided ever using the same series code for different things. Now it seems that this past practice has been ignored. As far as I can see, this is a big mistake at the USPTO.
The patent office people at the USPTO laid claim to series code 90 on July 1, 1981, assigning application number 90/000,001 to an ex parte reexamination request. For more than 38 years now, the patent office has been methodically assigning serial numbers one by one to ex parte reexams in series code 90.
For no reason that makes any sense, on June 13, 2020, the trademark office people at the USPTO tried to start using series code 90 for newly filed trademark applications, assigning application number 90/000,001 to a newly filed trademark application.
This mistake by the trademark people is starting to make trouble within the USPTO, as may be seen by this notice dated March 12 in the PAIR system:
Trademark is now using 90 series for new filings and those fees may be available in Reexamination Ex Parte proceedings when the Patent and Trademark serial [sic] numbers overlap. USPTO Teams are working to correct the issue.
(From a dictionary definition, we can see that the correct use of the term “serial” applies of course only to the part after the slash, because that is the part that increments “serially” meaning one by one.) But setting aside this poor usage of the term “serial”, there is no doubt that the trademark people were making a big mistake when they started doing this on June 13, 2020.
It looks like this could lead to no end of trouble. Suppose somebody sends in something to the USPTO by postal mail and it says “90/123456” on it. Does the USPTO employee in the USPTO mail room stick it into patent file 90/123456 or does the USPTO employee in the USPTO mail room stick it into trademark file 90/123456? You might say “oh it will always be clear from context which type of file it should get stuck into” but we all know that’s not realistic.
Similarly, if something arrives at the USPTO by fax, and it says “90/123456” on it, does the USPTO person in charge of incoming faxes stick it into the patent file or stick it into the trademark file? Same problem. Somebody mails or faxes form PTO-2038 to the USPTO (the credit card form) to pay a fee and it says “90/123456” on it. Does the credit card clerk at the USPTO post it to the patent file or to the trademark file?
And yes, in the Financial Manager system, if one fee got paid in patent file 90/123456 and another fee got paid in trademark file 90/123456, how are the people in Revenue and Accounting Management supposed to keep track of which file was which?
This strikes me as nuts.
By now I imagine it is way too late for the trademark office people to walk it back and renumber the new trademark applications filed on and after June 13, 2020 with a distinct series code.
Oh to have been a fly on the wall within the USPTO, and to see the consternation that surely arose on June 13, 2020 when the trademark people tried to assign that first trademark application number 90/000,001 to a trademark file, given that 38 years earlier the patent people had assigned application number 90/000,001 to a patent reexamination file. And on June 13, 2020, the revenue people had to try to figure out where to post the filing fee that the trademark applicant had just paid.
Boundy’s Laws of Naming (from a paper I published in ACM Software Engineering Notes back in the days when I was an engineer):
* never give two names to one thing.
* never give two things to one name.
* Names are meaningful and specific, and their length is proportional to their scope. A loop variable used only once in a two-statement loop may be called “i”, but a global variable that may be used anywhere in the program will have a long name that accurately describes its usage.
I know law firms and practitioners generally use their own identifiers for matters, but often, the folks in the docketing or accounting department ignore the law firm reference when an “Official USPTO Serial Number” is provided, since that number is supposed to represent the true matter in which some event occurred. Imagine a client’s surprise if they accidentally get billed a reexamination fee in connection with their trademark filing…
I filed a trademark application in November 2020. Shortly thereafter I attempted to file applications in other countries using the Madrid System. One of the nice features of the USPTO system is that you can auto populate the application using the US application or registration. The system would not recognize my application starting with the 90 code. I was told that there was a technical problem with 90 code applications and they were working on it. Took 4 months.
Eyes roll on floor laughing out loud. I saw this one coming as soon as somebody mentioned that trademarks were nearing the end of the 89 series code. Everybody at the USPTO who failed to see this coming should be cashiered.
I did once investigate the USPTO’s numbering system. Strictly, the 8-digit number is an “application number” or “control number” consisting of a 2-digit “series code” and a 6-digit “serial number”. But people who were around before the “series codes” came into general use when the USPTO was computerized always tended to use “serial number” indiscriminately for the 6-digit and 8-digit numbers.
Yes please see https://blog.oppedahl.com/?p=2661 where I discussed this back in 2017.