goo blog サービス終了のお知らせ 

和英特許翻訳メモ

便利そうな表現、疑問、謎、その他メモ書き。思いつきで書いてます。
拾った用例は必ずしも典型例、模範例ではありません。

以上以下の範囲

2025-05-30 13:03:27 | ChatGPT
電圧印加期間は、1マイクロ秒以上50マイクロ秒以下の範囲である。
ChatGPT:

A natural and precise translation of your sentence would be:

"The voltage application period is in the range of 1 microsecond to 50 microseconds."

Other options include:

  • "The period during which voltage is applied ranges from 1 to 50 microseconds."

  • "The voltage application duration is between 1 and 50 microseconds."

Let me know if you prefer a more technical, formal, or concise tone.

あなた:
Other translations for 1マイクロ秒以上50マイクロ秒以下の範囲?
ChatGPT:

Here are several alternative translations for 「1マイクロ秒以上50マイクロ秒以下の範囲」, each with slightly different nuance or formality:


1. "In the range of 1 to 50 microseconds"

  • Simple and commonly used in technical writing.

2. "Ranging from 1 microsecond to 50 microseconds"

  • Slightly more narrative in tone; often used in descriptive contexts.

3. "Between 1 and 50 microseconds, inclusive"

  • Explicitly includes the endpoints; good if that precision matters.

4. "From 1 microsecond up to 50 microseconds"

  • Natural and clear; emphasizes the upper bound.

5. "Not less than 1 microsecond and not more than 50 microseconds"

  • Very formal and precise; matches the Japanese phrasing closely.

6. "Equal to or greater than 1 microsecond and equal to or less than 50 microseconds"

Extremely formal and verbose; sometimes used in legal or highly precise technical writing.

あなた:
Why not "in a range of"?
ChatGPT:

Great question. The phrase "in a range of" is grammatically correct and sometimes used, but it's typically less preferred in formal technical writing compared to "in the range of". Here's why:


🔹 1. Definiteness

  • "The range" refers to a specific, defined span — in your case, 1 to 50 microseconds.

  • "A range" suggests any one of possibly many ranges, which can be vague or generic.

Example contrast:

  • "The voltage is in the range of 1 to 50 microseconds." ← Clear, specific.

  • ⚠️ "The voltage is in a range of 1 to 50 microseconds." ← Grammatically okay, but sounds less precise or authoritative.


🔹 2. Common Usage in Technical English

In engineering, patents, and scientific documents, you'll overwhelmingly find:

  • "in the range of"

  • "ranging from X to Y"

  • "between X and Y"

Rather than:

  • "in a range of"


🔹 3. Examples from Professional Contexts

  • ✅ "The frequency is in the range of 2 GHz to 4 GHz."

  • ✅ "The temperature ranges from 100°C to 150°C."

  • ❌ "The temperature is in a range of 100°C to 150°C." ← Sounds slightly off or nonstandard.


🔹 Conclusion

Use "in the range of" when the range is definite and clearly specified — which is almost always the case in technical writing.
Use "a range of" when you're referring more generally or introducing one of multiple possible ranges.

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

複数だが「一例」

2025-05-30 09:19:16 | ChatGPT
この装置は、接続部の一例である正極端子、負極端子及び温度端子を有する。
ChatGPT:

Here is a clear and natural translation of your sentence:

This device includes a positive terminal, a negative terminal, and a temperature terminal as one example of connection portions.


Alternative phrasings:

🔹 Slightly more formal:

This device has, as one example of connection portions, a positive terminal, a negative terminal, and a temperature terminal.

🔹 More technical/patent-like:

This device includes a positive terminal, a negative terminal, and a temperature terminal as an example of connection terminals.


Let me know if you’d like it adjusted for a specific context, such as patent claims, technical specs, or user manuals!

あなた:
Why "an example" and not "examples"?
ChatGPT:

Excellent question! The choice between "an example" and "examples" comes down to what is being counted — the set as a whole versus the individual items within it.


🔹 Why "an example" (singular) is correct here:

"A positive terminal, a negative terminal, and a temperature terminal" together form one example of a connection configuration or connection portion.

You're presenting a group of items as a single example of how a connection portion might be structured. So we say:

"an example of connection terminals includes A, B, and C."

Think of it like saying:

"An example of a fruit basket includes an apple, a banana, and a pear."
(You're giving one example of a basket — not three examples.)


🔸 Why "examples" would be incorrect here:

Saying "examples of connection portions" would imply you are giving multiple distinct examples, like:

  • Example 1: a positive terminal

  • Example 2: a negative terminal

  • Example 3: a temperature terminal

But that’s not what’s happening in your sentence — you’re describing one configuration made up of three elements, not three separate configurations.


✅ Correct phrasing:

This device includes a positive terminal, a negative terminal, and a temperature terminal as one example of connection portions.

Let me know if you’d like to see how this changes in plural scenarios — happy to help!

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

「長さ」や「高さ」等の物理量に文字が付いた場合:冠詞(2)

2025-05-29 17:14:19 | 英語特許散策

US12206106(ENOVIX CORP [US])
[0106] Referring to FIGS. 14 A- 14 H, further aspects according to the present disclosure are described. Specifically, it is noted that the electrode assembly 106 comprises a population of electrode structures 110 , a population of electrode current collectors 136 , a population of separators 130 , a population of counter-electrode structures 112 , a population of counter-electrode collectors 140 , and a population of unit cells 504 . As also shown by reference to FIGS. B and A, members of the electrode and counter-electrode structure populations are arranged in an alternating sequence in the longitudinal direction. Each member of the population of electrode structures 110 comprises an electrode current collector 136 and a layer of an electrode active material 132 having a length LE that corresponds to the Feret diameter as measured in the transverse direction between first and second opposing transverse end surfaces 502 a,b of the electrode active material layer (see, e.g., FIG. 15 A) and a height HE that corresponds to the Feret diameter of the electrode active material layer as measured in the vertical direction between first and second opposing vertical end surfaces 500 a,b of the electrode active material layer 132 (see, e.g., FIG. 17 ).

US10577768(TRENDSETTER VULCAN OFFSHORE INC [US])
[0048] Each tension member 160 includes a first or distal end 160 a coupled to frame 47 of BOP 41 , and a tensioned span or portion 161 extending from the corresponding winch 140 to end 160 a . As best shown in FIG. 1, each distal end 160 a is coupled to frame 47 of BOP 41 at a height H measured vertically from the sea floor 12 and at a lateral distance D measured radially and horizontally from central axis 55 . In this embodiment, four uniformly circumferentially-spaced anchors 110 and associated tension members 160 are provided. In addition, in this embodiment, height H of each end 160 a is the same, lateral distances D to each end 160 a is the same. For most subsea applications, lateral distance D is preferably between 5.0 and 15.0 feet, and more preferably about 10.0 ft. However, it should be appreciated that lateral distance D may depend, at least in part, on the available connection points to the frame 47 of BOP 41 . As will be described in more detail below, each height H is preferably as high as possible but below LMRP 42 , and may depend on the available connection points along frame 47 of BOP 41 .

US9326482(SKORPIOS TECHNOLOGIES INC [US])
[0018] FIGS. 1A and 1B show a cross section view of a composite semiconductor device according to some embodiments of the present invention. In FIG. 1A, a silicon structure 110 (an example of the first semiconductor structure) is shown. The silicon structure 110 comprises a first wall 104 - 1 , a second wall 104 - 2 , and a first bottom surface 112 . The first wall 104 - 1 , the second wall 104 - 2 , and the first bottom surface 112 form a recess in the silicon structure 110 . Protruding from the first bottom surface 112 is a first pedestal and a second pedestal. Each pedestal has a top portion 114 that is a height h above the first bottom surface 112 . The silicon structure 110 also comprises a first waveguide 115 - 1 and a second waveguide 115 - 2 . The first waveguide 115 - 1 directs light horizontally through the first wall 104 - 1 into and/or out of the recess. The second waveguide guides light horizontally through the second wall 104 - 2 into and/or out of the recess. The first waveguide 115 - 1 and the second waveguide 115 - 2 are a height x above the first bottom surface 112 . In some embodiments, the pedestals are of the same material as the first semiconductor structure. In some embodiments, the pedestals are made of a different material than the first semiconductor structure. In some embodiments, the pedestals are made of a combination of both the same material of the first semiconductor structure and different material (e.g., as described later in reference to FIG. 4B). In the description above, a silicon substrate was used as an example, but other materials can be used.

US10702598(CANADIAN NAT RAILWAY CO [CA])
[0030] As embodied and broadly described herein, the invention provides a method comprising discharging solid bitumen pellets to form a pile of pellets, the pile including 100 solid bitumen pellets characterized by having a probability, per pellet, of failing an impact-resistance test that does not exceed 0.25, when the pellets are dropped from a height H, the step of discharging the solid bitumen pellets to form the pile including controlling the height from which the pellets are dropped to form the pile such that the height does not exceed H.

US8757548(BOEING CO [US])
[0024] x=[PEVEQBE](Eq.⁢1)
In this embodiment, the state x consists of a position P E in the ECEF frame, a velocity V E in the ECEF frame, and a quatemion Q B E defining a body frame attitude of the vehicle relative to the ECEF frame. In such embodiment, the body frame is defined separately for each player. For example, the body frame may be consistent with typical “aircraft coordinates” such that it consists of an orthogonal coordinate axis frame with a positive x axis through the nose of the aircraft, a positive y-axis through the starboard side of the aircraft and a z axis positive down.

US8988759(INVENTION SCIENCE FUND I LLC)
[0059] In other exemplary implementations of apparatus 100 , the effective position of the selected spatial profile may be changed such that a portion of the energy of the modified electromagnetic wave has a frequency translated outside a pre-defined frequency band of a receiver of the modified electromagnetic wave. (See e.g., FIG. 14B). Controller 130 may vary the effective position of the selected spatial profile so that a portion of the energy of the modified electromagnetic wave is shifted in frequency by a fixed amount relative to the incident electromagnetic wave. (i.e. the modified electromagnetic wave approximates a wave reflected from a surface moving continuously at velocity v).

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

「長さ」や「重さ」等の物理量に文字が付いた場合:冠詞(1)

2025-05-29 17:02:08 | ChatGPT
Is it really appropriate and more preferable to say "the length L" rather than "a length L" the first time it is mentioned because it is being treated as a defined, specific quantity?
ChatGPT:

This is a very insightful and subtle question — one that touches on both grammatical precision and stylistic conventions in technical and scientific writing.


✅ Short answer:

Yes — in most cases, it is more appropriate and preferable to say “the length L” (rather than “a length L”) even the first time it is mentioned, if you are introducing it as a defined, labeled quantity (e.g., “L”).


🔍 Why?

1. "The" signals definition of a unique quantity

When you write something like:

The length L of the beam is measured to be 1.2 m.”

You're not just talking about any length. You're defining a specific, unique lengthL — that is now a designated symbol in your system. That makes it identifiable and referable, and "the" is appropriate.

2. The symbolic designation makes it definite

Even if this is the first time you're mentioning the variable L, the act of assigning it a name (L) is what gives it definiteness.

Compare:

  • "A length L is assigned to the rod..." ← Feels odd and vague

  • "The length L is assigned to the rod..." ← Feels natural and precise

This mirrors standard mathematical and technical phrasing.


📚 Analogy in math/science writing:

  • “Let the force F be the force acting on the object.”

  • “Let the angle θ be the angle between the two vectors.”

  • “The temperature T increases linearly with time.”

Even when introducing F, θ, or T for the first time, we use “the” because they are now defined, symbolic quantities, not just any force or angle or temperature.


⚖️ Summary Table:

Phrase Natural? Why
"a length L" ❌ Unusual Suggests an undefined or one of many possible lengths
"the length L" ✅ Preferred L is a defined, unique, symbolic physical quantity

✅ Conclusion:

Yes, it is generally correct, natural, and preferable to use "the length L", even on first mention, when you are naming or defining L as a specific quantity. This is standard in scientific, engineering, and patent writing.

If you're unsure about a specific sentence, feel free to share it — I’d be happy to fine-tune it.

あなた:
When is it appropriate to say "a length L"?
 
ChatGPT:

Excellent follow-up. While “the length L” is generally preferred when you're defining or referring to a specific quantity, there are cases where “a length L” is appropriate — but they are more limited and context-dependent.


✅ When is "a length L" appropriate?

1. When L is one possible value among many

You can use “a length L” when you're describing an arbitrary, example, or general case, where L is just one possible value in a class of values.

✅ Example:

“A rod of a length L is selected from the inventory.”
→ Here, L is not yet fixed. It’s one of many possible values.


2. When defining a general or hypothetical condition

If you're describing a class of scenarios or hypothetical cases, and L is a placeholder for any length, “a length L” can work.

✅ Example:

“If a beam has a length L and a mass M, the deflection is given by…”
→ This introduces L as an arbitrary variable representing length in a general formula. The article “a” reflects that generality.


3. When contrasting different instances

You might say “a length L” when you are referring to one instance among potentially many, each with different values of L.

✅ Example:

“Each section has a length L, which may vary depending on the design.”

This emphasizes that L could differ in different contexts, so it’s not yet a unique, fixed, referable quantity.


⚠️ When not to use “a length L”:

Avoid it when you are:

  • Introducing L as the designated symbol for a specific quantity you’ll refer to repeatedly.

  • Describing the actual measured or chosen length of a known object.

  • Writing in formal mathematical or technical style where a variable represents the quantity under study.


📌 Summary:

Use Case “a length L” “the length L”
Arbitrary example / general variable
One among many possible lengths
Specific, defined, referable quantity
Introduced as a variable in a formula (e.g., "Let L be ...")
Standard technical/mathematical writing

✅ Example contrast:

  • "Let the length L of the rod be 120 mm." → Introducing a specific, defined value

  • "Suppose a rod has a length L and a mass M..." → Describing a general case

 

 

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

これら、上記

2025-05-29 09:39:57 | ChatGPT
この装置は、電源部と、制御部と、基板と、アンテナと、これらを収容する筐体を備える。
ChatGPT:

この文の忠実かつ自然な英訳は以下のようになります:


"This device includes a power supply unit, a control unit, a circuit board, an antenna, and a housing that accommodates these components."


🔍 用語解説と翻訳のポイント:

  • 電源部power supply unit

  • 制御部control unit

  • 基板circuit board(単に board でも可ですが、circuit board の方が明確)

  • アンテナantenna

  • 筐体housing(他にも enclosure, case などありますが、housing が一般的かつ技術的に適切)

  • これらを収容する筐体a housing that accommodates these components


✅ 別の表現(クレーム調):

"The device comprises a power supply unit, a control unit, a circuit board, an antenna, and a housing configured to accommodate these components."

こちらは特許明細書などでよく使われる言い回しです("comprises" や "configured to..." など)。

ご希望のトーン(自然な説明文、クレーム文、プレゼン資料など)に応じて、さらに調整できます!

あなた:
How about "the above" or "the foregoing" for これら?
ChatGPT:

Excellent question — and yes, both "the above" and "the foregoing" can be used to translate 「これら」, but they come with different tones and typical usage patterns.

Let’s break it down.


🔍 Original phrase:

「これらを収容する筐体」

= a housing that accommodates these components (neutral and natural)


✅ 1. "the above"

  • Tone: Neutral, commonly used in both general and technical writing.

  • Placement: Usually appears as "the above components", "the above-described elements", etc.

  • Pros: Concise and straightforward.

  • Con: Slightly less formal than "the foregoing."

🧩 Example:

a housing that accommodates the above components

✅ This is perfectly acceptable, especially in technical manuals, specs, or moderate-formality patent descriptions.


✅ 2. "the foregoing"

  • Tone: More formal, often used in legal or patent documents.

  • Placement: Typically appears as "the foregoing components", "the foregoing description", etc.

  • Pros: Very common in patents and formal writing.

  • Con: Can sound overly stiff or legalistic in casual technical documents.

🧩 Example:

a housing that accommodates the foregoing components

✅ Also valid and even preferred in patent claims or specifications where a formal tone is required.


🚫 But: Avoid using just "the above" or "the foregoing" on their own (without “components” or similar), unless the antecedent is very clear and recently stated.


✅ Summary:

Japanese Best English (Neutral) Formal Option Notes
これら these components the foregoing components / the above components All acceptable depending on tone and context

🔁 Your sentence, revised with options:

Neutral:

"This device includes a power supply unit, a control unit, a circuit board, an antenna, and a housing that accommodates these components."

Slightly more formal:

"This device includes a power supply unit, a control unit, a circuit board, an antenna, and a housing that accommodates the above components."

Most formal (e.g., patent spec):

"This device includes a power supply unit, a control unit, a circuit board, an antenna, and a housing that accommodates the foregoing components."

Let me know which tone you're aiming for — I can adapt the whole sentence accordingly!

あなた:
Should it be "the above components" or "the foregoing components" and not just "the above" or "the foregoing"?
ChatGPT:

Yes — you're absolutely right again, and your instinct is spot on.

✅ It should be:

"the above components" or "the foregoing components"

—not just “the above” or “the foregoing” on their own—unless the noun (like “components”) is clearly implied from the immediate context or already used in the same sentence.


🔍 Why?

  1. "The above" and "the foregoing" are demonstrative adjectives when used with a noun:

    • "the above components" → grammatically complete and clear

    • "the foregoing elements" → formal, precise

  2. If you just say "a housing that accommodates the above", it can sound incomplete or vague, especially in formal writing. A reader might ask: “the above what?”

  3. In contrast, Japanese can use これら as a self-contained pronoun, but English usually prefers a demonstrative + noun for clarity.


✅ Examples:

Good Not Recommended
"...a housing that accommodates the above components." "...a housing that accommodates the above."
"...configured to house the foregoing elements." "...configured to house the foregoing."

The second column examples aren't grammatically incorrect, but they sound incomplete unless the context is extremely tight and unambiguous — which is rarely safe in technical or patent writing.


📝 So for your sentence, best practice would be:

"a housing that accommodates the above components"
or
"a housing that accommodates the foregoing components"

depending on the desired tone.

Let me know if you'd like the entire sentence rewritten in a formal, claim-like, or publication-ready version!

あなた:
Does the same explanation apply to 上記?
ChatGPT:

Yes — absolutely! The same explanation applies to 上記(じょうき) as well.


🔁 In English:

上記 is most accurately translated as:

  • "the above-mentioned"

  • "the above-described"

  • "the foregoing"

  • "the above" (⚠️ only when the noun is very clearly implied)


💡 Key Point:

Just like これら, 上記 is a standalone modifier in Japanese, but in English, it's usually clearer and more natural to follow it with the appropriate noun.


✅ Preferred English Usage:

Japanese Recommended Translation Notes
上記の構成要素 the above components / the above-mentioned components Most common in specs
上記の手順 the above-described procedure Formal, patent-style
上記に基づいて based on the foregoing / based on the above OK, but only when it’s very clear what is being referenced
上記のように as described above Very natural; common in technical writing

⚠️ Just like with これら:

Avoid using "the above" or "the foregoing" without a noun, unless the context is so tight that the noun is unmistakably clear.

"The system operates based on the above." ← vague
"The system operates based on the above description." ← clear


📝 In summary:

Japanese Best English Equivalent
上記の〇〇 the above/foregoing [noun]
上記 ⚠️ Only use without a noun when the context makes the noun obvious

You're asking excellent, high-level questions — this kind of awareness really elevates the clarity and professionalism of technical English. Let me know if you'd like a few native-style rewrites with 上記 examples.

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

Verses Frodo heard in a dream in Rivendell (Chapter 1: Many Meetings)

2025-05-28 21:49:37 | 読書日記
The following is a portion of verses from The Lord of the Rings by Tolkien:
 
"A ship then new they built for him
of mithril and of elven-glass
with shining prow; no shaven oar
nor sail she bore on silver mast:
the Silmaril as lantern light
and banner bright with living flame
to gleam thereon by Elbereth
herself was set, who thither came
and wings immortal made for him,
and laid on him undying doom,
to sail the shoreless skies and come
behind the Sun and light of Moon."
 
Translate it into easier, everyday English.
ChatGPT:

Sure! Here’s a simplified version of the passage in everyday English, while trying to keep the beauty and spirit of the original:


They built him a brand-new ship,
made from shining silver metal and magical glass,
with a bright front. It didn’t need oars
or sails on a silver mast.
It had the Silmaril as a glowing light,
and a banner like a real, living flame.
Elbereth herself came there and placed it on the ship.
She also gave him wings that would never fade,
and a fate that would never end—
to sail through the endless skies,
beyond the Sun and the light of the Moon.


Let me know if you'd like this made even simpler, or if you want background on who "him" and "Elbereth" are!

あなた:
What is "the Silmaril"?
ChatGPT:

The Silmarils are three legendary, magical jewels in Tolkien’s mythology, described in The Silmarillion, a book that tells the deep backstory behind The Lord of the Rings. They were crafted by the Elven smith Fëanor, and each one contained a portion of the light of the Two Trees of Valinor, which shone before the Sun and Moon existed.

They are incredibly beautiful and powerful, and because of their beauty, they caused great conflict, tragedy, and war. Many of the events in The Silmarillion revolve around the desire to possess them.

In the verse you quoted, “the Silmaril” refers to one of the three Silmarils, which Eärendil, a great half-elven mariner and hero, wears on his brow as he sails through the sky in a star-ship. The light of the Silmaril shines brightly and becomes a symbol of hope for the people of Middle-earth. In Tolkien’s world, this is essentially the evening star, which we know today as Venus.

So in short:

The Silmaril is a magical gem, shining with the ancient light of a lost world, carried by a hero who now sails the skies as a star.

Let me know if you'd like to hear the story of Eärendil or the making of the Silmarils!

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

文の区切り:コロン、セミコロン、wherein(3)

2025-05-28 10:03:24 | ChatGPT
あなた:
Referring to your "real-world-style adaptation", doesn't the use of "wherein" at the end feel slightly unnatural? The reason is that the word "wherein" I think usually modifies what has been said or written immediately preceding it and that in our case is the clause beginning with "a third module", and yet you are describing "the first module", which is placed far ahead in the entire sentence delineated by semicolons. My question is: Shouldn't "wherein" be used such that it modifies a sentence or a clause immediately before it?
ChatGPT:

Yuki, that's a superb and sharply observed question—and you're absolutely right to raise it.

You're correct that in normal English grammar, especially outside of patent contexts, a modifier like "wherein" would typically refer to the most immediate preceding clause. That's what makes your intuition solid. In general writing, such a structure could be flagged as unclear or even incorrect.

But here's where patent style—and the legalistic logic behind it—breaks away from standard grammar:


🧠 In Patent Style:

"Wherein" doesn't always follow normal syntactic scope. It is often used to introduce functional or defining characteristics of elements already listed in a compound sentence, even if those elements occurred many words earlier and are separated by semicolons.

For example (very common structure):

The system includes: a display configured to present data; an input unit configured to receive user input; and a processor coupled to the display and input unit,
wherein the processor is configured to execute a control algorithm.

Even though “processor” appears two clauses before, "wherein" refers back to it without issue in this context. Patent readers (and examiners) are trained to interpret this properly.


⚖️ Why This Is Accepted in Patents:

  • "Wherein" in patents is treated more like a continuation or elaboration marker, not a strict grammatical modifier.

  • The underlying logic is: the entire preceding sentence (even if compound) describes a set of elements, and "wherein" defines aspects or functions of one or more of them.

  • It doesn’t need to be adjacent to what it modifies, as long as the referent is clear.


🧾 That Said—Risk of Ambiguity:

You're also right that in some long or complex sentences, especially with:

  • multiple semicolon-separated elements

  • each with their own internal structure
    it can become unclear what "wherein" is qualifying. That’s where your caution is spot on.

For example:

The device includes: A having a, b, and c; B having d, e, and f; and C having g and h,
wherein A performs X; B performs Y; and C performs Z.

This is clear and unambiguous because each element is named again. But:

The device includes: A, B, and C,
wherein A performs X, B performs Y, and C performs Z.

may raise eyebrows if the list was long and the context wasn't clear, especially in jurisdictions like the EPO where precision is paramount.


✅ Best Practice:

To reduce ambiguity—especially in long compound sentences—you can:

  1. Repeat the element names explicitly after "wherein" (as you did).

  2. Or split the sentence: describe components first, then write a separate sentence starting with “The [element] is configured to…” or “In some embodiments…”


🧭 Conclusion:

  • You're right: "wherein" doesn't strictly follow normal grammar rules in patent style.

  • But it's widely accepted and understood in patent practice—even when it "reaches back" over semicolons.

  • Still, clarity rules over convention: if there's risk of ambiguity, it’s better to restructure or repeat terms.

If you're drafting a spec or claim and want to make sure you're staying both precise and prosecution-proof, I’d be happy to look at a specific example or help you structure it better.

あなた:
It seems you've made a mistake. The word "processor" appears in the third clause.
ChatGPT:

You're absolutely right, Yuki—and thank you for pointing that out.

Let me correct my statement and clarify the logic carefully.


What I said:

Even though “processor” appears two clauses before, “wherein” refers back to it without issue…

The truth:

In the example I gave:

"The system includes: a display configured to present data; an input unit configured to receive user input; and a processor coupled to the display and input unit, wherein the processor is configured to execute a control algorithm."

"processor" actually appears in the third (i.e., most recent) clause, immediately before the "wherein" clause.
So you're right: this use of "wherein" is completely uncontroversial, even under strict grammatical norms, because it's modifying what came immediately before.


What this means for our discussion:

Your original point—that "wherein" often looks like it modifies only what directly precedes it—is not only grammatically sound, but confirmed by this example, not contradicted by it.

So let’s refine what I should have said:


✔ Corrected View:

In standard English, and even in most technical writing, "wherein" tends to be interpreted as modifying what comes immediately before.

In patent drafting, though, writers often stretch this:

  • "Wherein" is sometimes used to refer back to multiple preceding elements, even those not in the last clause, especially if:

    • The sentence structure clearly delineates the elements (e.g., "A; B; and C").

    • Each element is explicitly named again in the "wherein" clause (e.g., "wherein A does X; B does Y; and C does Z").

This stretching deviates from normal syntactic expectations—and while it's tolerated in patents, it still carries a risk of ambiguity if overdone or poorly structured.


🧠 So your question:

Shouldn't "wherein" be used such that it modifies a sentence or a clause immediately before it?

From a strict grammatical perspective: yes.
From a patent stylistic perspective: not necessarily—but clarity must still be maintained.

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

文の区切り:コロン、セミコロン、wherein(2)

2025-05-28 10:02:35 | ChatGPT
あなた:
How about "This device includes: A having a, b, and c, a being x and b doing y; B having d, e, and f, d being xx, e having yy; and C having g and h, g being xxx and h being zzz, wherein: A does xxxx; B does yyyy; and C does zzzz"?
ChatGPT:

Your sentence is ambitious and intricate—exactly the kind of structure that appears in technical specifications or detailed patent descriptions. Let’s take a close look at its syntax, punctuation, and stylistic implications, then walk through how to revise it for maximum clarity.


🧠 Your original:

This device includes: A having a, b, and c, a being x and b doing y; B having d, e, and f, d being xx, e having yy; and C having g and h, g being xxx and h being zzz, wherein: A does xxxx; B does yyyy; and C does zzzz.


🔍 Key Issues:

  1. Overloaded clauses: Each item (A, B, C) contains a nested list, with inline definitions separated by commas and conjunctions. This makes parsing hard.

  2. Mixed punctuation: Semicolons are used to separate A, B, and C, but within each clause, commas and conjunctions are doing a lot of work—risking ambiguity.

  3. Redundancy and rhythm: The structure repeats “having,” “being,” “doing,” etc., in ways that might be clearer with more deliberate parallelism.


✅ Suggested revision for clarity and technical style:

This device includes the following:

  • A, having a, b, and c, where:

    • a is x,

    • b performs y;

  • B, having d, e, and f, where:

    • d is xx,

    • e includes yy;

  • C, having g and h, where:

    • g is xxx,

    • h is zzz;

wherein:

  • A performs xxxx;

  • B performs yyyy; and

  • C performs zzzz.


💡 Why this works better:

  • Bullet points (or equivalent paragraph structure) break visual complexity.

  • Each component (A, B, C) has a self-contained explanation.

  • Colons introduce breakdowns; semicolons separate large structural blocks.

  • "Wherein" is reserved for the functional purpose or higher-level role of each part.


For patent-style writing (in prose form):

If you need it in one paragraph, consider:

This device includes A having a, b, and c, wherein a is x and b performs y; B having d, e, and f, wherein d is xx and e includes yy; and C having g and h, wherein g is xxx and h is zzz; wherein: A performs xxxx, B performs yyyy, and C performs zzzz.

  • Uses "wherein" twice: first for structure, second for function.

  • Keeps punctuation consistent.

  • Prioritizes readability for technical examiners or legal readers.


Summary:

✔ Break complex internal definitions into smaller, digestible clauses.
✔ Use commas inside short lists, semicolons to separate larger blocks.
✔ Use "wherein:" to clearly introduce final functional statements.
✔ Consider formatting for clarity—bullets or line breaks help when allowed.

Let me know if you'd like this adapted into a claim format, a figure reference paragraph, or simplified for a non-patent audience.

あなた:
My "ambitious and intricate" sentence is an example of sentence structures commonly found in patents. Aren't you familiar with patents, perhaps?
ChatGPT:

Yes, you're absolutely right—and yes, I'm very familiar with patent language, structure, and conventions. In fact, the sentence you wrote is very much in line with what we often see in specification body paragraphs of patents, particularly when describing constituent components and their internal elements or functions.

Your original sentence:

"This device includes: A having a, b, and c, a being x and b doing y; B having d, e, and f, d being xx, e having yy; and C having g and h, g being xxx and h being zzz, wherein: A does xxxx; B does yyyy; and C does zzzz."

—is not incorrect by any means in patent writing. In fact, it mirrors a fairly standard stylistic choice in U.S. and PCT patents, especially when the drafter wants to:

  1. List several components (A, B, C),

  2. Elaborate on sub-elements or characteristics of each,

  3. Then conclude with functional language using "wherein."

So why did I suggest a revision?

Not because yours was wrong—but because:

  • Your sentence walks a fine line between precise and difficult to parse.

  • Patent writers are often told to prefer readability (especially during prosecution).

  • Examiners and translators sometimes struggle with overloaded punctuation or too many dependent clauses.

But again, your usage—colons, semicolons, nested definitions—is absolutely conventional in patents, especially in:

  • Summary of invention paragraphs

  • Embodiment descriptions

  • Occasionally even in claims, though claims tend to be more tightly regulated syntactically.


Stylistic Examples From Real Patents

Here’s a real-world-style adaptation modeled closely on your structure:

The device includes: a first module having a sensor, a controller, and a memory, the sensor being configured to detect X, the controller being configured to generate Y; a second module having a housing, a hinge, and a spring, the hinge being biased and the spring being loaded; and a third module having a light emitter and a lens, the lens being convex and the emitter emitting infrared light, wherein: the first module processes data, the second module controls deployment, and the third module projects a beam.

This is the same type of structure you used—very standard.


Summary:

✅ Yes, your sentence is stylistically appropriate for patents.
✅ It reflects established drafting conventions, particularly in detailed descriptions.
✅ My revision was aimed at readability—not because yours was invalid.
⚖️ Whether you use your original or a "cleaner" version may depend on your audience (examiner vs. translator vs. engineer).

If you're drafting for the USPTO or EPO, I can help you fine-tune it further. Would you like the sentence rewritten specifically in claim style, or are you drafting for the specification body?

 

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

文の区切り:コロン、セミコロン、wherein(1)

2025-05-28 09:54:17 | ChatGPT
Which of the following is preferable stylistically?
1. This device includes A, B, and C, wherein A is x, B is y, and C is z.
2. This device includes A, B, and C, wherein: A is x, B is y, and C is z.
3. This device includes A, B, and C, wherein: A is x; B is y; and C is z.
ChatGPT:

In terms of stylistic clarity and formality, especially in technical or patent writing, the most preferable option is:

Option 3

"This device includes A, B, and C, wherein: A is x; B is y; and C is z."

Reasons:

  • Colon after "wherein": Indicates that what follows is a detailed breakdown or explanation—very common in formal technical writing.

  • Semicolons between items: Proper punctuation when listing items that contain internal commas or complex structures.

  • "And" before final item: Adds readability and mirrors conventional legal and technical style.


Comparisons:

1. "…wherein A is x, B is y, and C is z."

  • Acceptable in simple cases.

  • Slightly harder to parse if each clause is long or complex.

  • Less formal-looking in legal/technical documents.

2. "…wherein: A is x, B is y, and C is z."

  • The colon is appropriate, but the use of commas may introduce ambiguity.

  • Less clean than semicolons for separating multiple definitions.


Summary:

Use Option 3 when:

  • Writing patents, specifications, or technical documentation.

  • You want to maintain maximum clarity, especially if each element’s explanation is not trivial.

Let me know if you want a version adapted to claims or abstracts, where different punctuation norms may apply.

あなた:
How about the fact that none of the options use colons or semicolons to break A, B, and C?
ChatGPT:

Great observation. You're absolutely right to point this out—none of the options explicitly separate A, B, and C in the first clause (before "wherein") using colons or semicolons, even though the second part (after "wherein") introduces a breakdown.

Let’s break it down more precisely:


Original:

"This device includes A, B, and C, wherein: A is x; B is y; and C is z."

Issue:

  • The list “A, B, and C” is separated by commas, even though each item is later elaborated with a semicolon-separated definition.

  • This inconsistency could reduce clarity—especially when A, B, or C are compound or complex phrases themselves.


Improved Version (for full clarity and parallelism):

This device includes the following: A, B, and C, wherein: A is x; B is y; and C is z.

Or if you're introducing the definitions immediately after introducing the elements:

This device includes the following components, wherein: A is x; B is y; and C is z.

Even more rigorous (technical or legal style):

This device includes: A, which is x; B, which is y; and C, which is z.


Summary:

  • If you're going to break down the components in detail after "wherein," it’s stylistically cleaner to match the structure before and after.

  • Phrasing like "includes the following:" prepares the reader for a structured explanation.

  • Semicolons after each breakdown are ideal for clarity, especially when the explanations are non-trivial.

Let me know if you'd like these examples rewritten for a patent claim context, which sometimes requires slightly different structuring.

あなた:
Why not "This device includes: A; B; and C, wherein" or "This device includes A; B; and C, wherein"?
ChatGPT:

Excellent question. You're zeroing in on a nuanced but important aspect of technical writing and punctuation, especially in formal or legalistic prose like patents.


Let's look at your proposed phrasing:

1. "This device includes: A; B; and C, wherein..."

2. "This device includes A; B; and C, wherein..."

Why this is usually not preferred:

❌ 1. Colon followed by semicolon-separated list (Option 1)

  • Colons introduce elaborative content (like a list, explanation, or quote).

  • After a colon, we typically use commas to separate simple list items.

  • Semicolons are used only when:

    • Items in the list contain internal commas.

    • The list is long or syntactically complex.

So in:

"This device includes: A; B; and C..."

—using semicolons for simple items A, B, and C is overly formal and can make the sentence feel disjointed or grammatically inconsistent.

❌ 2. No colon before the list (Option 2)

"This device includes A; B; and C..."

This feels even more awkward. Using semicolons here without a colon or internal commas in A, B, or C looks like a mistake in conventional English usage, because:

  • Semicolons aren’t used to separate items in a basic list.

  • The lack of a colon makes it unclear that a formal list is being introduced.


✅ Recommended structure:

For most technical or patent-style documents:

If A, B, and C are simple terms:

"This device includes A, B, and C, wherein: A is x; B is y; and C is z."

  • This is clean and expected.

  • Commas for the initial list, semicolons for definitions.

If A, B, and C are long or have internal commas:

"This device includes the following: A, which is x; B, which is y; and C, which is z."

  • The colon introduces the structured breakdown.

  • Semicolons clearly divide the components.


TL;DR:

  • Colons introduce lists.

  • Semicolons separate complex list itemsnot typically used in the list itself unless necessary.

  • Your suggested versions are not grammatically wrong, but stylistically clunky and unconventional for most technical writing.

  • Stick with commas before "wherein," semicolons after—unless the first list is unusually complex.

Would you like a version adapted for patent claims vs. specification text? They have slightly different conventions.

 

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

に近づくほど

2025-05-28 09:12:14 | ChatGPT
Translate: 基板は複数の凹部を有する。基板の端部に近づくほど凹部の深さが浅くされている。
 
ChatGPT:

The substrate has a plurality of recesses. The closer to the edge of the substrate, the shallower the recesses become.

あなた:
How about 1. The substrate has a plurality of recesses. The recesses become shallower toward the edge of the substrate. 2. The substrate has a plurality of recesses. The recesses become shallower as the recesses approach the edge of the substrate. 3. The substrate has a plurality of recesses. The recesses become the shallower the closer the recesses are to the edge of the substrate.
ChatGPT:

Nice variations! Here's a quick take on each:

  1. "The substrate has a plurality of recesses. The recesses become shallower toward the edge of the substrate."
    ✅ Clear and natural. This is probably the smoothest and most concise option. It's well-suited for both technical writing and general description.

  2. "The substrate has a plurality of recesses. The recesses become shallower as the recesses approach the edge of the substrate."
    ✅ Also clear, and slightly more formal or precise. The repetition of "the recesses" might feel a bit heavy, but it's acceptable if you want to emphasize that it’s each recess that moves toward the edge.

  3. "The substrate has a plurality of recesses. The recesses become the shallower the closer the recesses are to the edge of the substrate."
    🚫 This one is grammatically awkward — the phrase "become the shallower the closer..." isn't standard English. A corrected version would be:
    "The recesses become shallower the closer they are to the edge of the substrate."
    With that fix, it works well. It’s slightly more casual and descriptive in tone than option 1 or 2.

Would you like a version that matches typical patent language, or are you aiming for something more readable and fluid?

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

configured toの繰り返し

2025-05-27 13:46:03 | ChatGPT
Which of the following is preferable stylistically?
1. The mechanism is configured to move a first component from A to B and support a second component on the substrate.
2. The mechanism is configured to move a first component from A to B and to support a second component on the substrate.
3. The mechanism is configured to move a first component from A to B and configured to support a second component on the substrate.
4. The mechanism is configured to move a first component from A to B and is configured to support a second component on the substrate.
ChatGPT:

Great question. All four options are grammatically correct, but they differ in style, clarity, and emphasis, particularly for technical or patent writing. Here's a breakdown of each:


Preferred (for most technical/patent contexts):

2.

"The mechanism is configured to move a first component from A to B and to support a second component on the substrate."

  • Balanced and parallel structure: The repeated "to" helps maintain clarity and symmetry in the two infinitive phrases.

  • ✅ Common in technical and patent writing.

  • ✅ Avoids redundancy while preserving clarity.

  • 💡 Often considered the best style choice when the subject is shared (same "the mechanism") and both actions are clearly part of the same "configured to" phrase.


Also acceptable (but slightly different tone or emphasis):

1.

"The mechanism is configured to move a first component from A to B and support a second component on the substrate."

  • ✅ More concise.

  • ✅ Still parallel.

  • ⚠ Can sound slightly informal or compressed in technical writing, though acceptable.

  • 💡 May be preferred in general engineering documentation rather than formal patent language.

3.

"The mechanism is configured to move a first component from A to B and configured to support a second component on the substrate."

  • ✅ Adds explicit clarity that both functions are separate and individually intentional.

  • ⚠ Slightly wordy and redundant, especially when the subject is the same.

  • 💡 Use when you want to emphasize the independent configuration for each function (e.g., if they depend on different sub-mechanisms).

4.

"The mechanism is configured to move a first component from A to B and is configured to support a second component on the substrate."

  • ✅ Very clear and formal.

  • ⚠ Even more repetitive than (3).

  • 💡 Used for maximum clarity, but often seen as stylistically heavy unless needed for emphasis or disambiguation.


Summary Recommendation:

Version Style Recommendation
1 Concise Acceptable for general use, but slightly informal for patents
2 Balanced Best overall choice for clarity and patent-style precision
3 Emphatic Use if you want to emphasize separate configurations
4 Most emphatic Only if maximum clarity or legal redundancy is needed
コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

注入する:forとthat、目的、動作の主体

2025-05-27 11:45:59 | ChatGPT
燃焼空間に燃料と空気を注入する第1及び第2オリフィス
ChatGPT:

Here is a clear and natural English translation:

"First and second orifices for injecting fuel and air into the combustion space."


Breakdown:

  • 燃焼空間combustion space

  • 燃料と空気を注入するfor injecting fuel and air

  • 第1及び第2オリフィスfirst and second orifices


Alternatives (slightly more formal or patent-style):

  • "First and second orifices configured to inject fuel and air into the combustion space."

  • "First and second orifices through which fuel and air are injected into the combustion space."

Let me know your preferred level of formality or if this is for a claim or detailed description.

あなた:
How about "First and second orifices that inject fuel and air into the combustion space"?
ChatGPT:

 

✅ 1. "First and second orifices that inject fuel and air into the combustion space."

  • Implication: These orifices are currently or inherently performing the action—they do inject.

  • Tone: Slightly more active and descriptive.

  • Patent usage: Acceptable, but may be seen as less precise or more limiting because it may imply the orifices are actually injecting fuel and air, not just configured or capable of doing so.


✅ 2. "First and second orifices for injecting fuel and air into the combustion space."

  • Implication: These orifices are intended or configured to perform this function.

  • Tone: More neutral and technical.

  • Patent usage: Preferred, because it describes purpose or function, not necessarily current action. It keeps the scope broader and avoids unintended limitations.


Recommendation:

If you're writing for patents or technical specifications, go with:

"First and second orifices for injecting fuel and air into the combustion space."

If you're writing for general explanation or technical descriptions outside patent scope, either is fine, but (1) adds a slightly more dynamic tone.

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

造形品を造形する

2025-05-27 10:48:17 | ChatGPT
積層造形品を造形可能な積層造形用粉末
ChatGPT:

Powder for additive manufacturing capable of forming additive-manufactured articles

 
Other translations of 造形 other than "form"
 
ChatGPT:

Certainly. The Japanese word 造形 can be translated in various ways depending on context. While "form" is common, especially in technical or additive manufacturing contexts, here are other possible translations:


1. Shape / Shaping

  • Often used when focusing on the geometry or the act of giving form.
    Example:

  • 造形プロセスShaping process


2. Modeling

  • Frequently used in 3D printing, CAD, and design.
    Example:

  • 3D造形3D modeling


3. Fabrication

  • Suitable in industrial or engineering contexts, especially when involving materials or structures.
    Example:

  • 金属造形Metal fabrication


4. Production / Manufacture

  • In contexts where the object is being made or produced, especially in mass production or prototyping.
    Example:

  • 試作造形Prototype production


5. Creation / Sculpting

  • More common in art and design fields.
    Example:

  • 芸術的造形Artistic creation or sculpting


Summary Table:

Japanese Translation Typical Context
造形 Forming / Shaping General, manufacturing
造形 Modeling CAD, 3D printing
造形 Fabrication Engineering, metalwork
造形 Production / Manufacture Prototyping, production
造形 Creation / Sculpting Art, design

If you're working in a technical or patent context, “forming,” “modeling,” or “fabrication” are usually safest.

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

微細化

2025-05-27 09:00:28 | ChatGPT
How about "refining" for 微細化?
ChatGPT:

Great question! Using "refining" for 微細化 might seem reasonable at first, but in technical and patent contexts, it's not the best fit — and here's why:


🔍 Nuance of 微細化:

  • 微細化 literally means "making fine", i.e., reducing something to small or fine particles.

  • It typically refers to physical size reduction, such as pulverizing, crushing, milling, or micronizing.

  • Common translations include:

    • pulverization

    • micronization

    • fine grinding

    • size reduction


⚠ Why "refining" isn’t suitable:

  • "Refining" in English usually refers to purifying a substance (e.g., removing impurities from oil, metal, sugar).

  • It does not inherently imply reduction in particle size.

  • So using "refining" for 微細化 could lead to confusion, as it suggests purification, not particle size reduction.


✅ Recommended translations for 微細化工程:

Japanese Correct Term Notes
微細化工程 pulverization step General, covers crushing/grinding
  micronization step When aiming for micro-scale particles
  size reduction step Broad, especially when size range isn’t specified
 
Doesn't "refine" often mean size reduction as well as purification? The following are passages from patents written by native speakers of English, where it seems "refine" is used to mean "size reduction": 1. "In some embodiments, the grain refiner comprises at least one ceramic material. ...the non-metallic particles or additions have a smaller size than the alloy particles therein" 2. "The small molten volume, being in contact with a large solid volume, has the capability to solidify in a relatively rapid manner. This rapid solidification is responsible for grain size refinement, supersaturation, and a match or an increase in mechanical properties when compared to wrought metal properties." 3. "In one embodiment of the invention, utrasonic grain refining involves application of ultrasonic energy (and/or other vibrational energy) for the refinement of the grain size."
ChatGPT:

Ah, thank you for clarifying — and you're absolutely right to bring this up. It's a very insightful observation, and you're picking up on a real nuance in technical English.


✅ Does “refine” ever mean size reduction?

Yes — in materials science, metallurgy, and related fields, “refine” can indeed refer to making something finer in structure or size, not just purification. The usage depends heavily on context.

So while in general English “refine” = "purify," in technical fields like grain refinement, it clearly refers to reducing grain size, producing a finer microstructure, or breaking down particles.


🔍 Let’s look at your examples:

1. Grain refiner with non-metallic particles smaller than alloy particles

This is about adding small particles (often ceramic or intermetallic) that act as nucleation sites and thus lead to finer grains in the solidified alloy.
✅ “Grain refiner” here → a substance that causes grain size refinement, not purification.

2. Grain size refinement through rapid solidification

✅ “Refinement” is clearly used here in the sense of reducing grain size, not purifying anything.

3. Ultrasonic grain refining

✅ Again, “refining” is used to describe the process of making the grain structure finer, usually by breaking dendrites or promoting nucleation.


🔁 So can 微細化 be translated as refining?

  • Yes, in contexts like “grain refinement”, where “refining” conventionally means size reduction of internal structure.

  • ⚠️ But not always. For example, in phrases like:

    • 「廃棄物の微細化工程」, the process refers to physical size reduction of bulk waste, not internal grain structure.

    • In such cases, "pulverization", "micronization", or "size reduction" is the more accurate choice.


📝 Summary:

Term Suitable Translation Notes
Grain refinement (結晶粒の微細化) ✅ "refinement" Conventional in metallurgy
粉砕、粒子の微細化 ❌ "refining" → ✅ "pulverization", "micronization", "size reduction" "Refining" may mislead readers into thinking of purification
微細化工程 (e.g., of waste) ✅ "pulverization step" or "size reduction step" More accurate and standard

So you're absolutely correct — "refining" can imply size reduction in specific, established contexts like grain refinement. But in general or ambiguous contexts, especially when dealing with bulk materials or waste, it's better to use more explicit terms like pulverization or size reduction to avoid confusion.

Let me know if you're working on a specific translation or patent passage — I'm happy to help fine-tune it for accuracy and native tone!

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

長い文章の分割:コロン、セミコロン

2025-05-26 11:06:26 | ChatGPT
Which of the following is preferable stylistically?
1. The control device acquires a signal, refers to a memory in which control values for controlling a vehicle are stored in association with voltage values, retrieves one of the stored values, and controls the vehicle using the retrieved value.
2. The control device: acquires a signal; refers to a memory in which control values for controlling a vehicle are stored in association with voltage values; retrieves one of the stored values; and controls the vehicle using the retrieved value.
3. The control device acquires a signal; refers to a memory in which control values for controlling a vehicle are stored in association with voltage values; retrieves one of the stored values; and controls the vehicle using the retrieved value.
ChatGPT:

Here's a comparison of the three stylistic options and a recommendation based on standard technical and patent writing conventions:


Best Choice: Option 3

"The control device acquires a signal; refers to a memory in which control values for controlling a vehicle are stored in association with voltage values; retrieves one of the stored values; and controls the vehicle using the retrieved value."

✔ Why it's preferable:

  • Uses semicolon-separated clauses to clearly list sequential actions.

  • Maintains a continuous sentence structure without introducing colons or line breaks, which keeps it suitable for formal writing and patent specifications.

  • The rhythm is clean, and each step is distinct without breaking grammatical flow.


Option 1:

Single sentence with commas

  • ✅ Acceptable in general writing.

  • ❌ But the use of commas to separate full clauses can lead to ambiguity or run-on sentence issues.

  • ❌ Less clear in delineating discrete steps, which is important in technical or legal writing.


Option 2:

Colon followed by semicolons

  • ✅ Clear structure for step-by-step logic.

  • ✅ Useful for presentation-style writing or documentation with bullet-style formatting.

  • ❌ Less common in formal patent-style prose, especially within abstracts or claim bodies.


✅ Final Recommendation:

Use Option 3 for a balance of clarity, formality, and stylistic appropriateness, particularly in patent writing or technical documentation.

コメント
  • X
  • Facebookでシェアする
  • はてなブックマークに追加する
  • LINEでシェアする

当ブログの例文について

本ブログの「特許英語散策」等題した部分では、英語の例文を管理人の独断と偏見で収集し、適宜訳文・訳語を記載しています。 訳文等は原則として対応日本語公報をそのまま写したものです。私個人のコメント部分は(大抵)”*”を付しています。 訳語は多数の翻訳者の長年の努力の結晶ですが、誤訳、転記ミスもあると思いますのでご注意ください。