The analysis of grammar in the frame challenge is valuable, but the 'laziness' and 'energy conservation' answers are doubtless correct ... but unsupported claims: and Mari-Lou is correct in not dignifying a question 'basically asking for guesses and opinions, which is off-topic' with an 'answer'.

@EdwinAshworth For supported references, please see the linguistics research of Lauren Collister at the University of Pittsburgh, as well as that of Lauren Fonteyn at the University of Manchester. Both study language use in informal digital contexts and the complex social dynamics we see driving evolving conventions there, including in particular this matter.

The OP's relatively casual formulation of the question ('Why do some people . . . ?) was somewhat ambiguous. Interpreted one way, it could have been given a very simple, short answer: because there is a widely accepted convention of writing this way (in certain settings). It is only if the question is interpreted as a question about the causes of the emergence of the convention, that it needs a more elaborate answer.

@jsw29 Would that really have worked in our format? An answer that short would run the risk of being auto-flagged by the system as Very Low Quality, getting sent to the VLQ Review Queue for likely downvoting and possible deletion, and inviting a post notice reading “This question needs detailed answers, including citations and an explanation of why an answer is correct.”

Although you are undoubtedly right that the convention 'was common in electronic texts even before the Internet', it was probably given a boost by the widespread use of smartphones, whose virtual keyboards, combined with small size, makes it more time-consuming to shift between lowercase and uppercase.

@tchrist, agreed, the question deserves a proper answer on this site only if it is interpreted the second way; interpreted the first way, it should have arguably been closed. My point was only that the formulation of the question was open to both interpretations.

@jsw29 I did indeed myself initially close it as POB. But then I saw that the OP's public profile mentions that they are in Japan, whose hiragana and katakana are both unicameral, and mulling over the matter a bit, I decided that it would be possible to answer in a way that would be worthy of our site and therefore reopened it, awaiting answers from others. Then I saw people taking potshot answers in comments which saddened me since it would likely lead to no actual answers, so I set about answering it myself. I encourage others to please do so as well if they have other perspectives to offer.

Such text-speak style was established usage in a profession I trained in a half century ago, and wasn't new then. We would write "hbw in cb ff b/s" and we all knew it meant "Half brick wall in common bricks with a fair face both sides", which is what would appear in the finished document.

I grew up with NYC subway ads like "u cn gt a gd jb w hi pa!" for steno courses. We all already know informal writing shortcuts are nothing new (nor very complex). I'd point out the "They are intended to be more like those quickly written notes that you pass to your friends in class" in this answer. I would think that social media writing would be a bit more...social. Perhaps that's why I stay away from it.

"but the 'laziness' and 'energy conservation' answers are doubtless correct" – correct in some cases but not all. when i'm typing chat messages (or, in this case, a Stack Exchange comment) on my phone, my keyboard app automatically capitalizes the beginnings of sentences and the pronoun "i," so i have to perform extra taps in order to make them lowercase. there are various reasons why i do this, and this answer gives a good overview of the reasons

Essential reading: Because Internet by Gretchen McCulloch

Syntax is more than "word order". And syntax and punctuation are related if not inseparable. Leaving aside its relationship to speech, punctuation marks syntactic boundaries. The whole point of introducing punctuation in the 16th century was to be a marker of syntax ...

... CamGEL has a chapter on punctuation, and they say that "These (punctuation marks) serve to give indications of the grammatical structure and/or meaning of stretches of written text. The punctuation marks are all segmental units of writing ... There are, however, various non-segmental features which can serve the same kind of purpose as the punctuation marks... And while the end of a sentence is indicated segmentally by a punctuation mark (...), the beginning of a sentence is indicated non-segmentally by capitalisation of the first letter."

@ishtar You're still talking about the encoding technology, not about the sound-based language that this technology is encoding. Otherwise we would have capital letters and punctuation — and italics and bold face and blackletter and cursive — in our actual speech, which we do not. Writing is encoding aspects of speech that you can hear, intonation and pauses and such, but which when put on the printed page would otherwise be lost. Fluent speakers can be completely illiterate; they do not need to know how to spell things or punctuation them or capitalize them to speak.

I have used English as an analogy for programming languages, but the opposite can be done here. A semicolon at the end of a statement in the C-family of programming languages is part of their syntax, whether one would read them out loud or not. Missing a semicolon at the end of a statement would cause a "syntax error". Because we don't pronounce full stops in speech doesn't mean they're not there.

"As with spelling, capitalization cannot be “grammatically wrong” because it is about writing, which is merely technology for encoding actual language, not the language itself." First of all, there is some sense in which written language is not merely an encoding of spoken language, but a separate language that shares a lot of properties with spoken language. Sometimes those two languages can diverge significantly. For instance, written Arabic is not merely an encoded form of spoken Arabic. And with Chinese, you even have one written language that encodes several different languages.

Secondly, written language can encode grammatical differences, so it doesn't make sense to say that it can't be grammatically wrong. For instance, "led" is the past tense of "lead". Someone who writes "lead" when they mean "led" is encoding a grammatically incorrect form of the word. How is "The international maneuvering of the late 18th and early 19th century lead to the outbreak of WWI" not grammatically incorrect?

@Acccumulation Sorry, but spelling mistakes are not grammatical errors. They are technological transcription errors. Otherwise nothing anybody wrote using alternate, older, or different writing conventions would be grammatical, and that is clearly not true. Linguists do not consider orthography to be part of grammar. Encoding errors are simply that. Native speakers who never learned to read and write still produce grammatical speech. They just don't know how to write down what they said. Writing is technology. Speech is not.

@Mari-LouA Well, that's kind of my point. When I write the pronoun "I" and starts of sentences in lowercase, my purpose is not to save time; my purpose is to express myself by using a writing style that I feel fits me. The fact that that writing style violates the standard rules of English writing is part of the reason why I use it—I don't want to be seen as someone who follows all of the rules all of the time.

@SophieSwett good luck with writing answers on EL&U capitalising will-nilly words, and suggesting to learners that writing the pronoun I in lowercase (i) is about style and personality. When I see an answer on EL&U and ELL not using conventional spelling, typos, inexistent capitalisation, and poor punctuation, my opinion of that answer is considerably lower compared to someone who cares enough to follow norma, standardl writing styles. Texts, I suppose are a different kettle of fish.

@Mari-LouA: Correct: part of the purpose of wishing to appear as a rule-breaker, is wishing to forgo the approval of those people (including you in this case) whose high opinion is withheld from rule-breakers. For a while this was called "punk", but of course it has many other incarnations :-). I would say, though, that over on English Language Learners, it probably is a service to questioners to model rule-following.

@SteveJessop I'm certain that my opinion matters not one jot to Sophie, I was merely reminding the person (and you) that EL&U is an English language site, and answers which show little attention to spelling, and capitalisation (of which I've seen many) do not earn praise. As for my finding ee cummings insufferable, you assume wrongly. But I believe that text messages are not works of literature or poetry. If you're going to stick to lowercase, and want to make a point, be consistent (note that Stack Exchange was capitalised)

@Mari-LouA I try to always use a style that's appropriate for the medium and the situation. On Stack Exchange, that means that I do use conventional spelling and capitalization (aside from my initial comment, which I wrote in the very style that it was describing, for rhetorical effect). In chat rooms on Discord with my friends, I use a different style, in which the rules "capitalize the first letter of a sentence" and "capitalize the pronoun 'I'" are dropped (but other capitalization rules remain in place).

My chat room messages are "not works of literature or poetry" indeed, but by using the style that I do, I accomplish what I intend to accomplish.

@Mari-LouA: I didn't assume your opinion of EE Cummings / ee cummings. I just meant it as another example of someone who would disapprove Sophie (because of the specific non-standard use, rather than deprecating any and all non-standard typography). I don't really understand why you connect Sophie's post with "showing little attention to ...", because as far as I could see it shows very close attention. In fact it contained fewer typos than your post complaining about (among other things) typos, which is the inevitable tragic irony of correcting people's typography on the internet!

@SteveJessop unfortunately, there are only 5 minutes in which to correct typos in a comment. And also, I am very much aware of my short comings but I seek to do my best.

@tchrist If the person writing "lead" intends lēd, is that ungrammatical? Does the grammaticality of the writing depend on the state of mind of the writer? "Otherwise nothing anybody wrote using alternate, older, or different writing conventions would be grammatical" There are grammatical variants, and there are orthographic variants. I don't see a contradiction.

As a person who was actually alive at the time, I can confirm that such lowercase typing was commonplace back in the 70s and 80s among early computer users.

Let me try another way. There is more to syntax than the spoken (phonetic) form. Grammars have ellipses, null constituents, "hidden levels", et cetera. Sign language is non-spoken, but has syntax. I see no reason to say that punctuation does not belong with syntax because it is not spoken.

“If it’s the same words in the same order, it has the same grammar” — the classic (if somewhat puerile) counter-example to this erroneous statement being "I helped my Uncle Jack off his horse" Putting the sentence entirely in lower case changes the grammatical class of one of the nouns into a verb, and completely changes the meaning of the sentence along with its grammar.

@Chronocidal Your two sentences are pronounced differently; the intonation pattern is completely different, which is a signal that they have different syntactic structures. We must therefore conclude that they are not the same words. We normally pronounce words subtly differently within the overall clause structure when they're doing different jobs syntactically: time flies like an arrow versus fruit flies like a banana. Our primitive writing system only rarely captures any of the myriad intonation nuances and cues that we nonetheless rely upon in the spoken language for understanding it.

@tchrist And use of a different/incorrect word is a grammatical error. Thus, you prove my point and disprove your own. Your obsession with an apparent stance that grammar is something that you seem only to recognise as existing in spoken language is highly discriminatory. Are you suggesting that mute and/or deaf people are incapable of either grammar or errors in it?

@Chronocidal Thank you for sharing your knowledge; before you explained this I had been unaware that road signs in all capitals were “not grammatically correct”: STOP, EXIT, YIELD, ONE WAY, WRONG WAY, DO NOT ENTER, SHARP CURVE AHEAD, SPEED LIMIT 60 MPH, CROSS TRAFFIC DOES NOT STOP, OVERHEIGHT VEHICLE PULL OVER, METRIC SIGNS IN USE ON INTERSTATE 19, ACCIDENT AHEAD TUNNEL CLOSED ALT ROUTE US 6, DRIVER PULLED FROM CRASH BLOSSOMS. Have you told the highway department?