Email Subjects

Email subjects. What are they? What do they communicate? What are the conventions? Who decided these things? How have they changed over time? Where are they headed? Are they obsolete?

I wrote an email today that was somewhat formal, in that it was more formal than:

subject: none

body: Whattup! How’s the slow cooker? Still burning roasts?

More along the lines of:

subject: Thanks

body: Person,

Wanted to say thanks again for taking the time to speak with me today,…

But I hesitated on the subject line before sending: should it be capitalized? lowercase? more than just “thanks”?

According to the Internet Message Format, aka email spec, apparently the subject field in an email is actually just a header. And optional at that! Not that this changes much, but we see it’s just a string in a hash of headers.

From the Online Etymological Dictionary, we get the following entry:

subject (n.)

early 14c., “person under control or dominion of another,” specifically a government or ruler, from Old French sogit, suget, subget “a subject person or thing” (12c., Modern French sujet), from noun use of Latin subiectus “lying under, below, near bordering on,” figuratively “subjected, subdued,” past participle of subicere, subiicere “to place under, throw under, bind under; to make subject, subordinate,” from sub “under” (from PIE root *upo “under”) + combining form of iacere “to throw” (from PIE root *ye- “to throw, impel”). In 14c., sugges, sogetis, subgit, sugette; form re-Latinized in English 16c.

Meaning “person or thing regarded as recipient of action, one that may be acted upon” is recorded from 1590s. Grammatical sense is recorded from 1630s, from Latin subjectum “grammatical subject,” noun use of the neuter of the Latin past participle. Likewise some restricted uses in logic and philosophy are borrowed directly from Latin subjectum as “foundation or subject of a proposition,” a loan-translation of Aristotle’s to hypokeimenon. Meaning “subject matter of an art or science” is attested from 1540s, probably short for subject matter (late 14c.), which is from Medieval Latin subjecta materia, a loan translation of Greek hypokeimene hylē (Aristotle), literally “that which lies beneath.”

Particularly interested in the translation of the Greek hypokeimene, “that which lies beneath.” Can we then think of email subjects as that essence that lies beneath our meandering email or, perhaps, what the heart yearns to say in our emails that don’t say enough.

Sigmoid Curves

S-Curves:

The beautiful S-Curve. A curve we encounter daily, revealed itself tonight as an optional output from machine learning algorithms.

“The function was named in 1844–1845 by Pierre François Verhulst, who studied it in relation to population growth. The initial stage of growth is approximately exponential; then, as saturation begins, the growth slows, and at maturity, growth stops.

The logistic function finds applications in a range of fields, including artificial neural networks, biology (especially ecology), biomathematics, chemistry, demography, economics, geoscience, mathematical psychology, probability, sociology, political science, linguistics, and statistics.”

What more proof does one need; look at that lineup!

Photogrammetry

I’m utterly speechless. And it’s due to photogrammetry.

In a land far, far away, I was a land surveyor out West. In those heady times of shooting lasers, I become familiar with the idea of point clouds as representations of geometric space in coordinate form. Per our execution with lasers and reflective prisms, the resolution of our models were decided by how many measurements we took. For a big job, this might be in upwards of 100-300 points. Imagine hitting the corner of a building with a laser – all corners – then a few shots along the sides to confirm that walls weren’t billowing, or beams sagging. 14 hours later, you had 100, 200, 300 points in 3D space that would form a point cloud.

This was amazing.

Well, technology marches on. Sometime in the last 5-7 years, I stumbled on a project called Photosynth, that offered the ability to upload a series of overlapping photos to create a point cloud, where points were determined by parts of the images it determined were overlapping and the same point in space. Incredibly, it would deduce a 3D space from 2D images. This was absoluteley jaw-dropping at the time, and still is. It was also shortlived. The project was through Microsoft and the University of Washington, but with the proliferation of quadcopter drones and GPUs and more efficient and widely available processing of images, I’m assuming this technology was suddenly highly profitable.

I will keep this short for now, as the hours loom long, but truly consumer grade software has started to emerge that allow photogrammetry rendition of 3D models from images; much like Photosynth, but exportable to formats well understood in the 3D modeling space. Some is proprietary software like Photoscan, but other promising open source options like OpenDroneMap are hot on their heels.

My first photogrammetric project, a pair of 2+ year old Nikes with more miles than I can count. A particular fondness I have for old shoes; effortlessly they are by your side (or under your feet) in life moments of small to large; what else propel and protect like they do.

It is clearly quite rough, but, this was 12 images, taken quickly, from my phone. 12 images. From a phone. In poor lighting.

Undoubtedly, more to come. Much, much more. In fact, we are soon to meet with someone at Wayne State whom I understand to be doing this very thing for cultural artifacts. Now, how do we preserve, and provide discovery and access for these beauties…