Engineering & Standards

Entries in this section cover engineering, computing, measurement and technical standards. Each gives the substance of one document and points at the body that issued it.

They are not a syllabus. Each was a document that other work cited at this address, so the section follows what people actually reached for: a wartime suppressed pistol, a wingtip experiment, a packet-capture tool, an XML vocabulary for strategic plans, a method for measuring what standards are worth, and a procurement argument about a transport aircraft. The summaries below exist so a reader arriving from a citation can tell within a paragraph whether this is the document they were sent to find.

What the entries cover

The OSS silenced pistol: the High Standard HDM. The HDM is a .22 Long Rifle target pistol fitted with an integral suppressor, developed for the Office of Strategic Services, and it is the most widely used clandestine firearm of the Second World War — one of the few weapons of the period still in service four decades later. Its design succeeded because it accepted a constraint that louder weapons could not. A suppressor works on muzzle blast; it cannot do anything about the crack a supersonic projectile makes in flight. The HDM bleeds gas out of a ported barrel early, keeping the round subsonic, and accepts the reduced velocity that follows. Everything else about the design is downstream of that decision.

Multiple winglets and induced drag reduction. Induced drag is the price of lift. A finite wing generates a pressure difference between its lower and upper surfaces, air spills around the tip from high pressure to low, and the trailing vortex that results is energy shed into the wake. For a transport aircraft in cruise this accounts for roughly a third of total drag, and a larger share in climb. Splitting the tip into several small surfaces splits the circulation rather than concentrating it in one vortex, and the measurement is honest about the size of the result: the gain is real in the tunnel and small when weighed against the structural and certification cost of carrying it.

tcpdump: packet capture, BPF filters and reading TCP flags. A packet sniffer reads traffic passing a network interface, and the same capability serves opposite purposes — an attacker harvesting credentials and an administrator finding out why a connection fails. tcpdump is the standard command-line tool for the second case, and the filter expression is what makes it usable, because an unfiltered capture on a busy interface produces more than a person can read. The practical core is the flag pattern: a SYN with no reply, a SYN answered by RST, and a completed handshake followed by an application-level failure are three different faults, and the capture distinguishes them in a way that no application log does.

Strategy Markup Language: a history of StratML. Every public body publishes a strategic plan, and they contain the same elements — mission, vision, goals, objectives, measures, responsible parties. They are published almost universally as PDF or prose HTML, which makes them readable by people and opaque to software. StratML is an XML vocabulary for those elements, developed through AIIM and issued as an ANSI standard, so that plans can be aggregated, compared and checked against reported performance rather than read one at a time. The history is worth reading for what it shows about adoption: the technical problem was the easy part.

Measuring the economic benefits of standards. Standards are widely assumed to be economically beneficial and are rarely measured. ISO developed a methodology to change that — a structured way of quantifying what standards contribute to an individual organisation, applied across a series of company case studies in different sectors and countries. The method works bottom-up inside a single organisation: map the value chain, identify which functions standards actually touch, select the value drivers for each, and put a monetary figure on them. The unit of analysis is also the limit. It measures the firm, not the economy, and the part it structurally cannot see is the network effect — much of what a standard is worth accrues between organisations rather than within any one of them, and that share is probably the larger one. A firm-level result is therefore a floor rather than an estimate, which is the opposite of how such figures are usually quoted.

The C-130J Super Hercules cost-effectiveness dispute. The C-130J is the current production model of an aircraft first flown in 1954. The Center for Defense Information, through its Straus Military Reform Project, argued from Air Force budget data that it was cost-ineffective relative to the older models it replaced. The argument is worth examining because the reasoning generalises: the whole disagreement is about the baseline. Measured against the fleet it actually replaced, the J model's changes — new engines and propellers, a two-crew flight deck — buy a certain amount. Measured against what else the money could have bought, they buy something else entirely. Neither side is doing arithmetic wrong.

A thread running through several of them

Three of these entries are about turning an assertion into something that can be checked, and they fail or succeed at it in instructive ways. ISO's method exists because "standards are beneficial" was an article of faith with no number attached, and it produces numbers at the level of a firm. StratML exists because a strategic plan asserts goals and measures in a format that no machine can compare with the next body's goals and measures. The procurement argument is what happens when two parties both have numbers and disagree about the baseline they are measured from, which is the disagreement that numbers alone never settle.

A second thread is narrower and worth stating because it is so easily lost in summary: two of these entries report a real gain and then say how small it is. The winglet work does it about induced drag, the transport-aircraft fact sheet does it about the J model's improvements. Work that qualifies its own result is the work most often cited for the headline and not the qualification.

Where the section boundary falls, and where it does not

Two of these entries sit here on a judgement that could reasonably have gone another way. The HDM pistol is as much a piece of wartime history as a piece of engineering, and it is filed here because what makes it worth a record is a design decision about gas and velocity rather than an operational history. The aircraft dispute is procurement policy, and it is filed here because the argument is conducted entirely in engineering and budget terms. Neither is a claim about what the subject really belongs to; both are decisions about where a reader would look first, and the index lists every entry on one page regardless of section.

What the plates show, and where there is none

Four of the six entries carry a drawn plate; two do not. A plate is drawn where the content is structural or numeric and a diagram states it more precisely than a sentence, and nowhere else — a figure that restates a paragraph is padding with a border around it.

The suppressor entry carries a section through the integral unit, because the argument is about where gas goes and in what order. The winglet entry carries what splitting a tip does to the vortex, and what that buys against a plain tip. The tcpdump entry carries the three handshake patterns a capture shows and what each one localises — network, service or application. The C-130J entry carries what the J model changed alongside the baseline the cost critique measured against, which is the single fact the dispute turns on. The StratML and ISO entries carry none, because both are about method and process, and neither has a shape.

Who holds the originals

The publishers here span four quite different kinds of body, and the kind determines how a citation behaves over time. A standards organisation — ISO, or ANSI through AIIM — issues documents with permanent designations, and those designations survive revisions, which means a citation without a version or year is ambiguous in a way it would not be elsewhere. A university laboratory such as Georgia Tech publishes through conference papers and technical reports, where the stable identifier is usually the report number rather than the title.

A policy institute such as the Center for Defense Information publishes advocacy research: properly sourced, openly argued from a position, and to be read as such. A trade periodical such as Small Arms Review, and an independently published technical author, sit outside the academic indexing systems entirely — the work is real and locating it means going to the publication rather than to a database.

Going from a citation to the document

Each entry names its publisher and, where the original carried one, its reference. Search on the number first, the title second, the author last. For the standards documents in this section, add the year or the edition: a designation alone identifies a family, not a version.

Where a document exists in several forms — a conference paper and a later journal version, a draft standard and its published edition, a fact sheet and the testimony behind it — they are not interchangeable, and pagination will not match across them. This register records what each document was and who published it; it does not reproduce the documents themselves.