Notices by tom jennings (tomjennings@tldr.nettime.org)

@JessTheUnstill

LORD OF THE RINGS, for example? SHEESH.

@The_Caretaker

Americans generally disdain public transportation as a handout "for poor people", think it's dirty (and sometimes it is, because...) and so stations are installed in the cheapest locations not usable for other desirable things. I'm not exaggerating for effect.

@argv_minus_one

That would fail: binary file transfers toss all 256 combos at the port. Compression assured the spread.

In practice, delay +++ delay worked perfectly. I don't think it happened even once to me or anyone I talked to (very large number).

Also the worst that could happen was disconnect; the host could talk to the modem but from the line side, nada.

Modem control and recovery was a deepish art in the FidoNet days. They weren't that determinate when things Went Wrong. And there were like 20 or 50 different modem types in use. Some were just awful. Some were wildly complex. Some could not be used as a BBS incoming.

Getting serial ports with consistent control lines as per anything like rs232 was awful. The damn apple machine had zero handshake lines. Things were wildly inconsistent then.

@counternotions

I think it really means, "I want money, and I'll do anything to get some"

@mos_8502

That would have left me out, and more than half the people I know who wrote significant code.

I agree with the goal, but I doubt it would have/had and effect on eg. Google, openai, whatever.

I have no schooling at all.

Someone here pointed out that most of the major TLDs (top level domains eg com edu org etc) are somewhat at the whim of US govt and corporations, and to consider getting a TLD managed elsewhere (added redundancy).

I'm looking for something relatively generic like .info, alas that seems to be managed by a US corp.

I'd rather not choose a country code without good reason. Or on a registrar in an authoritarian country.

Any suggestions? Tried finding a list of TLDs' hosting countries but I'm not making good searches I guess.

My registrar is in Germany.

@anthk

~/public_gopher

Won't that effect the URL?

> images subdir link

I can't visualize what you mean.... lynx doesn't do images, how would one render an image stored on remote port 70 onto the local screen?

I can configure clients for my *own* use OK, but for the hypothetical other-user, if they have to manually configure a viewer for each file type, yeah, that won't fly. Unless I'm missing somethign, which is often the case?

@screwtape

OK I have a tentative gopher server up at gopher://squirrel.bleeve.me:70

#gopher

I'll stick with this for a while, but I'm not sure I'm ready to give up the ability to display images, or see PDFs. Those are just too valuable.

And it hangs, handing off to the h... URL: handler. The lynx client locks up.

I can't seem to get the server to serve from any root except /var/gopher, which means I'm editing sudo'd, which is gross, and I'm tolerating it only because I'm testing (and assume I'm making some stupid config misteak).

But the experiment has me rethinking presentation of content. I'm now writing simple, compliant HTML5, and doing pulldown menus in css. I'm using basic tables, lists, and some wrap around images. I may simplify this further. I could ditch the menus going forward.

But I don't think I give up inline images. They're just too meaningful.

I blurted out, a few days ago, "what about plain HTML on port 80?" instead of 443 and security etc.

WHAT I MEANT WAS, what if we were to stick to 80, for social, not technical reasons?

Comparing 80 and 443, utterly ignoring for a moment how and why they are different --

Port 443 is fraught with security problems, most of the internet seems to be (...) 443, its where big assholes like amazon reside, logins and encryption, complexity and overhead.

Port 80 is... boring as shit. Most of the vintage computing resources are located on it. They're super low overhead, text, images, files just available for use. They (are, can be) easily indexed. Read by anyone.

No one will do ecommerce on port 80. This is a good thing, but also a different discussion.

Gopher... the presentation is simplistic, but with a bit of artistry that arises natually when you work in it. That there's no tools could be just temporary.

But other than contrariness, its mostly security through obscurity. (Lots of obscurity.) But it does seem very unlikely that the creeps will start crawling port 70 any time soon, so that obscurity may be significant.

I'm not convinced that port 443 provides usable protection from personal tracking and characterization. Maybe it did once; I don't think it does not now. The Fediverse protocols have come to a similar conclusion.

I wish I could serve the same content on 80 and 70.

I wish I could translate gopher to html.

I wish I could have one "secure" page for ecommerce, or maybe not, and just farm that out at the tiny scale.

Back in 1986, when I started work at Apple (OS/ROM group) I had to sign the usual pile of (paper) forms, one of which was that anything that I work on is Apples. I told them I would have to have my lawyer look at it first. This took them by sufficient surprise that I was accepted as employee, but no one ever later asked me for the signed agreement.

Upon quitting less than a year later -- I hated it there, did poorly, and wasn't particularly liked I suspect -- they caught it, and withheld accrued vacation or something unless I signed it. But by then all my work was on the past and sorted out so I signed it.

I did get to work beside Bill Atkinson for 15 or 30 minutes, that was cool. As newbie I was supposed to find some bug in the timer manager. I didn't. I wow write the code that caused a floppy fo eject if pit press apple key+ 1. No, that is definitely not significant.

That was my last straight-up industry job. I fkn hated it lol.

What if we put <zerowidth>FUCK YOU FUCK EVERYTHING FUCK FUCK</zerowidth> in every HTML file?

Or javascript that computes pi to a billionaire's places, between the closing </BODY> and closing </HTML> tags where no sensible browser or human will see?

Hmmm or maybe a rethink; how to rig up a server that parses HTML documents for serving in gopherspace. Now that would be win.

@kdriscoll

No but it does look interesting. Mark pointed it out too. Id only skim read about it.

Conversion of my fairly vast HTML is the main issue, scripts to do the work would probably be involved and radically change presentation so I dunno.

I wish there was some way to signal via port or protocol that my HTML site has nothing but html and css.

@ariadne

Di.fm is a small scale streaming service, a corp but not publicly traded. I don't know their relationship to the performers. It's narrowly focused on dance music. I've been a member for a decade. I'm part owner too, lol, they did a fundraiser once by selling shares, I own 100.

No ads of course. You can + and - the stream with like dislike. Lots of channels.

They serve as contrast to the big scumbags.

@artologica

I tell our dogs, "those are professional animals. They do that for a living."

They don't listen.

My website is very old (1994) and varied in popularity between a little and full-on obscurity. There were times in there when I wanted more clicks etc but now I'm like wut; it's where I put things I want to persist, often written for some target purpose, link posted somewhere, then it remains for reference or whatever. This is all fine.

But the last few years, even for stuff on it that is rarely found elsewhere, and that google would dredge up (FIELDATA character code meanings, Librascope, whatever) has essentially disappeared; search engines mostly stopped indexing stuff that isn't commerce.

(I had a two year period when I did all the SEO shit, figured out it was all bullshit, and stripped it all out; Google owes me ~ $80 for ad income (lol) for those two years but they don't write checks under $100. "Policy".)

So moving to gopher space will cost me (sic) nothing.

In fact it will save me (sic): no security.

There is no way a mere mortal such as I can win at the encryption war. That is a corporate game. I will never win but I will never play a game that's rigged.

All of my websites are pure 100% HTML and CSS. I was fine with http: until the EFF thing. I hate SSL as it makes hosting data dependent on corporate entities, de facto.

So I won't lose any clicks (sic) if I move to gopher. You can get to it with lynx or links. And maybe I'll ditch cloudflare ( Iwas gonna revert to lets encrypt, but etc) and https entirely and just tell people, my site is a mere document, it can't track you. Well it could but it doesn't; look at the page source.

Screw this, I've had it, I'm moving back to gopherspace.

I think I'm serious.

https://metacpan.org/pod/Apache::GopherHandler

This was well after my desqview time, somwhere in the mid 90s i'd switched to 386/bsd. but it looks very nice.

I can attest the multitasking of earlier desqview was top notch.

https://lunduke.substack.com/p/desqviewx-the-forgotten-mid-1990s

OK sheesh, THIS was a super-sweet spot in my (admittedly ancient) computing platform history: DESQView and QEMM.

https://en.wikipedia.org/wiki/DESQview

I mistakenly remembered it as OS/2, which I did run around that time (88, 89, 90) but no, it was desqview.

I'm fine without a mouse, it turns out. I've been writing a bunch of CP/M sofware and old workflows came back (hint: its barbaric, I don't fully recommend it).

Gonna put a web page up about the CP/M thing soon. Got my version 2 boards from JLCPCB, awaiting a delayed delivery from Digikey.

#retrocomputing

@liaizon

I was outspokenly opposed to any commercialization of FidoNet and actively sabotaged efforts to do so at a big IFNA convention, and signed on to some legal shenanigans with the packing of votes and other things. That umm probably didn't help.

Thanks for mentioning me!