Getting the benefits of DOCX filing while avoiding the risks

There is a way to get nearly all of the supposed benefits from the validations provided in USPTO’s DOCX filing paths, while avoiding the ponderous professional liability risks of those DOCX filing paths. 

The USPTO says, to anyone who will listen, that a reason you should subject yourself to the professional liability risks of filing in DOCX format is so that you can get the benefit of validations performed by USPTO’s black-box DOCX-to-PDF rendering engine. In this blog article I will tell you how you can get the benefit of all of these validations (to the extent that the validations are beneficial) without having to assume any of the professional liability risks of filing in DOCX format.

The USPTO says that the DOCX e-filing system “detects common errors, such as formatting errors, and provides instant feedback to prevent unnecessary delays in processing your application.”  Let’s start by enumerating and evaluating the (supposedly beneficial) validations that the USPTO DOCX e-filing system provides.  After that, I will tell you how you can get all of those validations while avoiding all of the professional liability risks of doing a DOCX filing.

If after this evaluation you feel that you really would benefit from those validations, then by all means make use of them in the way I describe at the end of this blog article.

What are the “common errors, such as formatting errors” that the DOCX e-filing system will warn you about?  The USPTO details them here (archived here).  I enumerate them below, grouped by topic.

Claims.  A first category of warnings relates to the claims of a patent application.  The warnings are:

    • The claims are not numbered with positive integers.
    • The claim status identifier is not valid.
    • The claims contain an improper dependency (missing or canceled claim).
    • The claims contain duplicate claim numbering.
    • The claims contain an improper dependency (with a claim depending on itself).
    • The claims contain a claim with an improper dependency as it does not depend on a previous claim.
    • The claims contain a claim which references back to another multiple dependent claim.
    • A claim does not end with a period.
    • A claim does not begin with a capital letter.
    • A claim contains more than one sentence.
    • The claims contain a multiple dependent claim that does not use the alternative form.
    • No claims were detected.
    • The claims contain a claim that references two sets of claims to different features.
    • The claim contains more than one period at the end of the claim.
    • A claim number was detected that appears to end with a comma rather than period.
    • Missing or out of order claim number detected.

I think many practitioners would share my reactions to this list of warnings.

First, if it is a few minutes before midnight and if I need to get a same-day filing date, it is actually a bug, not a feature, for USPTO’s e-filing system to throw any of these claim warnings in my face.  Every one of these supposed flaws in the claims can be fixed later.  If I were to drop what I am doing to fiddle with a claim that fails to end with a period, probably midnight would come and go, and I would lose my filing date.  Yes, years ago to get a filing date, a patent application was required to contain at least one claim, but that was years ago.  Now in 2024 the only kind of patent application that is required to have a claim in it to get a filing date is a PCT patent application.

Second, I think it is pretty clear the real reason for these validations is not to help the filer.  The real reason for these validations is to make it so that Examiners do not encounter these errors.  A cynic would say that this is to set it up for some future time when USPTO management will try to cut back on the allowed amount of time that an Examiner is allowed to spend, per patent application, in the examination process, based upon the successful culling-out of such errors before the application reaches the Examiner’s desk.

Third, I think most practitioners seldom make these claim errors.  In the rare instance that a practitioner does make any one of these errors, it is fixable later.  To the extent that there is any category of applicant that makes such claim errors, and might conceivably benefit from such warnings, it is pro se filers.  But in my view, a pro se filer has no business doing a DOCX filing in the first place, given all of the risks involved.  The pro se filer that wants the benefits of such warnings can, in any event, get the benefits of such warnings while avoiding all of the DOCX risks, as described below.

Paragraph numbering.  A second category of warnings relates to paragraph numbering.  The warnings are:

    • The specification paragraph numbering is not in the proper format.
    • The specification paragraph numbering is not consecutive.
    • Sequential paragraph numbers were not detected.
    • Paragraph numbering is missing from the specification.
    • Duplicate paragraph numbering has been detected.

I think many practitioners would share my reaction to this list of warnings.

First, if it is a few minutes before midnight and if I need to get a same-day filing date, it is actually a bug, not a feature, for USPTO’s e-filing system to throw any of these paragraph-numbering warnings in my face.  If I were to drop what I am doing to try to insert paragraph numbering into a patent application that I am getting ready to file, probably midnight would pass and I would lose my filing date.

Second, nothing in 35 USC or 37 CFR requires paragraph numbering!   The MPEP merely suggests the use of claim numbering.  If you were to look through the most recent five hundred patent applications that I have filed at the USPTO, that led to an issued patent, you would find that … wait for it … not one of those patent applications had paragraph numbering in it.

The Abstract.  A third category of warnings relates to the Abstract.  The warnings are:

    • Word count in abstract section is over 150 words.
    • The abstract contains more than one paragraph.

The alert reader will have no difficulty predicting what I am going to say about these warnings.  First, who cares?  I certainly do not care.  You do not even need to have an Abstract to get a filing date!  And if it is a few minutes before midnight and if I need a same-day filing date, I absolutely promise you I do not want to be spend the limited number of minutes between now and midnight reading this warning and having to arrive at a judgment call that I do not care about the warning.  To the extremely limited extent that I care at all about this, I will wait to see if the Examiner gripes about it, and only if that happens will I worry about fixing it later.

The cynic will conclude that this, too, is being injected as a barrier in the e-filing path not to benefit the customer but to permit shaving off a minute or two in the permitted amount of time for an Examiner to write that first Office Action.

Let’s look at the remaining four warnings.  Can you stand the suspense?  Are the remaining warnings important?  Do they represent mistakes that practitioners often make?  Here are the remaining four warnings:

    • No page numbers were detected.
    • This document contains no recognized section headings.
    • The page size of this document exceeds Letter or A4.
    • Non-embedded font error.

Let’s work through these one by one.

No page numbers were detected.  It is true that 37 CFR requires page numbers.  But, as I describe in this blog article, the requirement for page numbers is an anachronism, dating from the old days when patent files were on paper and there was a risk the physical pages might get dropped onto the floor and would need to get sorted back into proper order.  In the world of PDF filing (which began more than 20 years ago) there is no opportunity for pages to get “dropped on the floor”.  Page numbers absolutely do not matter in this modern e-filing world of EFS-Web and, now, Patent Center.  Nonetheless, every now and then after we receive a Notice of Allowance, we will receive a notice from the Issue Branch telling us that we must provide page numbers (see blog article).  Never mind that even without the supposedly crucial page numbers, somehow the Publication Division manages to do the 18-month publication.  Never mind that through the entire examination process, the Examiner is able to examine the application and figure out that the invention is patentable, all without the supposedly crucial page numbers.  But now when it is almost time to pay the Issue Fee, for the very first time, some clerk at the USPTO feels the need to enforce the rule that says there must be page numbers.

So yes, in a weird way I can see how it is sort of helpful to flag a to-be-filed patent application if it lacks page numbers.  But once again, if it is a few minutes from midnight, I absolutely do not want this source of delay thrown in my face.  The lack of page numbers will not jeopardize my filing date, and, does not (apparently) get in the way of a successful 18-month publication, and does not (apparently) get in the way of a successful examination of the application for patentability.  But if the warning serves to reduce the work burden on the Issue Branch, then it is all worthwhile, I guess (/s).

This document contains no recognized section headings.  Neither 35 USC nor 37 CFR requires section headings.  The MPEP suggests section headings.  But section headings are not required by any law or rule.

If I am working on getting a patent application filed before the stroke of midnight at the USPTO, it will be a bug, not a feature, to have a warning thrown in my face about the absence of section headings.  The absence of section headings does not lead to loss of a filing date, and the absence of section headings is not (properly) a bar to patentability.  To the extent one were to feel the need to address this, it could be done some time after the filing date has been secured.

The page size of this document exceeds Letter or A4.  The fact is that once a patent application has entered the USPTO’s DOCX workflow, the page size no longer matters.  (In much the same way, the putative margins in a filer’s DOCX file are wholly irrelevant.)  The USPTO’s DOCX workflow leads to the USPTO controlling the page options and the margins to the exclusion of any page size or margins that the filer’s original word processor file might have indicated.  I cannot figure out why anyone in USPTO’s DOCX developer world actually cares about this, since all they are doing is manipulating text into other font sizes and formats.

Non-embedded font error.  This putative benefit of choosing a DOCX filing path is sort of a red herring.  What the author of this part of the USPTO web page about DOCX is really trying to say is that for twenty years now, filers who file in PDF sometimes try to upload a PDF that depends upon some font that happens not to have gotten embedded into the PDF file.  The idea here is that if the USPTO forces users not to file in PDF, then by definition it is impossible to upload a PDF that depends upon some font that happens not to have gotten embedded into the PDF file.

What is disingenuous about this supposed benefit of filing in DOCX format (that one cannot inadvertently attempt to upload a PDF that depends upon some font that happens not to have gotten embedded into the PDF file) is that it only presents itself in filing path 2 above (DOCX and no PDF).  But that filing path is the maximal-malpractice path.  No informed practitioner would ever select that path anyway.

The second-worst of the three filing paths is “DOCX plus protective PDF” and … wait for it … that filing path has the same old risk of the non-embedded font error as the first path (legacy PDF).

Anyway, filers have been following the legacy PDF path for two decades now, and have somehow worked out how to create PDF files that avoid the non-embedded-font error.   Anybody who prints to PDF using CutePDF automatically avoids that error.  Anybody who uses Libre Office and exports to PDF automatically avoids that error.

I can’t quite shake my skepticism that the lack of an embedded font actually gets in the way of the USPTO receiving a PDF.  I note, for example, that anybody who wishes can upload any PDF (including a PDF with some non-embedded font) to the ePCT system, and the ePCT system has no difficulty figuring out how to receive that PDF file.

How to get the benefits with none of the risks.  Which brings me to what I promised to tell you, namely how to get the benefits (such as there are) of the USPTO DOCX validations, while avoiding any of the profound professional liability risks of filing in DOCX.

There are actually two ways to get the benefits.

Commence the DOCX filing path, and have the self-control to refrain from clicking “submit”.  Yes, if you feel you cannot do without the supposed benefits of the USPTO DOCX validations, one way is simply to commence the DOCX filing path, receive the warnings, and then … cancel the submission.  You can then get the benefit of the warnings, and then you can start a new e-filing submission, following a less risky filing path such as “legacy PDF”.

 Use demo mode.  The USPTO developers have provided a “sandbox” in which a person who wants to play with the DOCX filing path may do so in a safe and harmless environment.  To get the supposed benefits of the USPTO DOCX validations, just play in the sandbox, upload your DOCX file, receive the warnings, and then … leave the sandbox.  You can then get the benefit of the warnings, and then you can start a new e-filing submission, following a less risky filing path such as “legacy PDF”.

One Reply to “Getting the benefits of DOCX filing while avoiding the risks”

Leave a Reply

Your email address will not be published. Required fields are marked *